Analysis report

Property Values

Data from R package wooldridge (CRAN), published under https://diafunc.com/licences/view-only.

Target
Rprice
Analyzed
2026-11-07 13:16 UTC

Summary

The model predicts Rprice and relies most on Larea, which accounts for 76% of its reliance. It explains 29% of the variation in Rprice, with the strongest baseline, a straight line through the data, explaining 52%. This means the model did not beat the baseline, finishing 23 points behind on variance explained. Raising Larea from 7.30 to 7.81 raises the model’s output by about 18,918 on average, with everything else as observed.

Because the model did not outperform a straight line through the data, individual predictions are not more useful than that yardstick. The model uses specialist functions in a formula of moderate length, and its errors differ between groups for Nearinc, Y81, and Y81nrinc. The next step is to validate the core relationships or seek simpler or stronger formulas before relying on the model for decision-making.

Formula

Rprice = 77700 − 0.9 × lerp(Larea, 7.35, Land)Rprice=77700−0.9×lerpLarea,7.35,Land
Rprice = 77700 − 0.9 × lerp(Larea, 7.35, Land)
2 terms · additively decomposable

Held-out diagnostics

TargetQualityErrorR²MAESearch rowsGeneralizationRows
Rprice16.54%70.99%0.290123372.2257generalizing64
  • Rprice · Quality 95% CI 0% to 29.04%
  • Rprice · Error 95% CI 60.36% to 92.77%
  • Rprice · R² 95% CI 0.07231 to 0.3964
  • Rprice · MAE 95% CI 19090.4 to 28136.3

Concerns

WARNING model

Held-out residuals carry a systematic offset: the model over-predicts by 8,106 on average

Across 64 held-out rows, predictions sit 8,106 above the actual values on average, against a residual standard deviation of 28,555: the offset is 28% of the residual spread. The offset is small relative to the residual spread but consistent in direction, so individual predictions lean the same way. Account for it wherever predictions are added up or compared with a threshold, and check on the next run whether it persists.

WARNING model

Residual variance grows with predicted value: model accuracy degrades on higher-magnitude predictions

Across 64 held-out rows split into thirds by predicted value, the residual standard deviation in the top third is 35,436, against 15,318 in the bottom third, 2.31 times as large. The bottom third holds predictions up to 78,031 and the top third predictions from 91,424. Relative to the predictions, the typical miss is 21% of the value in the bottom third and 35% in the top third, 1.65 times as large. The residual spread is 2.31 times as large at the high end: noticeable, not severe. Predictions at the high end deserve caution; a log transform of the target may tighten them.

WARNING model

The model does not beat a straight line through the data on `Rprice`

The model explains 29% of the variation on the held-out rows, where a straight line through the data explains 52%: 23 points less. The interval of the margin includes zero: on this many held-out rows the two are not told apart. The formula adds little over the reference predictor; whether it is worth its complexity depends on the use. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had.

INFO model

3 columns track `Rprice` almost completely

3 columns each tell almost everything about Rprice on their own, 1 of them as exactly as a copy would. The closest are Lrprice (Spearman's rho = 1 (Pearson r = 0.959)), Lprice (Spearman's rho = 0.955 (Pearson r = 0.921)) and Price (Spearman's rho = 0.955 (Pearson r = 0.952)). The coefficients read the same for a strong input and for a leak. A column that is known before Rprice exists is a strong input, and the quality stated in this report is real. One that was computed from Rprice or recorded after it makes a formula that will not hold on new data. The model reads none of them, so its predictions do not depend on the answer. Each column needs its own answer: confirm it as known before the target, or leave it out and run the analysis again.

Complete list (3)
ColumnAssociationReadingModel
LrpriceSpearman's rho = 1 (Pearson r = 0.959)Matches like a copyDoes not read it
LpriceSpearman's rho = 0.955 (Pearson r = 0.921)Almost determines itDoes not read it
PriceSpearman's rho = 0.955 (Pearson r = 0.952)Almost determines itDoes not read it

Analysis

The report opens with the Headline and the Concerns, followed by the Discovered Relations when the analysis has targets and by Questions & Answers when questions were asked. Three sections carry the detail: Model presents one model per target, Implications states where the models hold and what to weigh before relying on them, and Data describes the table as supplied, column by column, for reference.

Generated 2026-11-07 13:16 UTC, analysis fe39ec60-48b7-11f1-8087-d378640fe787, 1 target, time budget 1 h 0 min, compute budget 500 credits, 321 rows. The Provenance & Audit chapter holds the full record.

Validity

Validity: valid. Nothing in the data or the formula speaks against relying on this result.

How to read this

Whether this result is one to rely on. Quality says how closely the formula follows the data; validity says whether that closeness means anything. A formula that only restates its target, or that reads a column which already holds the answer, can reach a high quality and still say nothing new about the data.

What it found

Rprice starts from a base of 77,700 and rises with Larea and Land.

  1. Larea carries most of the model. Larea accounts for 76% of what the formula explains and Land for 24%.
  2. Rprice rises with Larea. With the other inputs at typical values, the formula gives 49,078 at Larea 6.62 and 124,000 at 8.53. The change is not even across the range.
  3. Baths is not in the formula. Its correlation with Rprice is 0.69. Rows with more of it have more Larea (correlation 0.71), and Larea is in the formula.
  4. The formula does worse than a straight line. The formula explains 29% of Rprice; a straight line explains more, 52%. That was measured on 64 rows it never saw. The formula typically misses Rprice by 23,400.

Summary

What to do next: Too few held-out rows to judge coverage, largest gap 5.00 percentage points.

6 chapters, each headed by what it found.

S.1 Larea carries most of the formula

What the formula relies on. Shares of the formula: Larea 76%, Land 24%.

Measured on 25 of the 257 rows the model was found on, which the search set apart for it. On so few rows the order of the inputs is indicative, not settled. The model's output rises with Larea on every one of the 8,320 steps between observed values. With Land the model's output rises on 5,824 and falls on 2,496 of the 8,320 steps between observed values. Moving Larea from 7.30 to 7.81, its first and third quartile, with every other input as observed raises the model's output by 18,918 on average (by between 788 and 251,126 on single rows). These are measurements on the rows the search trained on, inside the observed ranges; outside them nothing is claimed.

Table 5.1.3 · Which inputs matter most
InputShare of importancePermutation importanceSensitivityResidual correlationOccurrencesModel's output as the input risesChange, first to third quartile
Larea75.76%0.883736333.7-0.20621Rises (8320 of 8320 steps)+18918.2 (7.297 to 7.81)
Land24.24%0.28270.39590.19671Rises and falls (rises on 5824, falls on 2496 of 8320 steps)+5277.8 (16,935 to 45,973)
Rows the search trained on, measured
130
N rows
25
Inputs the formula does not read
Age, Agesq, Area, Baths, Cbd, Dist, Intst, Ldist, Lintst, Lintstsq, Lland, Nbh, Nearinc, Rooms, Wind, Y81, Y81ldist, Y81nrinc, Year

The formula decomposes into 2 additive terms: clean separable structure. Each curve below shows one term's response to its free inputs, with the other inputs held at the training median. Steep curves carry the prediction; flat curves are decorative.

Additive
Yes
Rows the search trained on, measured
130
Terms in the formula
2
Notation
77700 − 0.9 × lerp(Larea, 7.35, Land)
Target
Rprice

S.2 A straight line does better here: R² 0.29 against 0.52

How well it holds. The formula explains 29% of Rprice; a straight line explains more, 52%.

How this was checked

  • Measured on 64 rows the formula never saw (20% of 321); the other 257 built it.
  • Compared on the same rows against always guessing the average, always guessing the median and a straight line.
  • Intervals from 1,000 bootstrap resamples of those rows.

R² of 0.29 on 64 rows the search never saw (95% interval 0.0723 to 0.396): weak. The model captures only a small part of the variance; either the relevant inputs are missing, or the target depends on interactions the formula did not recover. Mean absolute error is 23,372 on the target's scale.

Error
70.99% (95% CI 60.36% to 92.77%)
MAE
23372.2 (95% CI 19090.4 to 28136.3)
N
64
Training rows
257
Quality
16.54% (95% CI 0% to 29.04%)
R²
0.2901 (95% CI 0.07231 to 0.3964)
Rows found on
257
Target
Rprice

The model does not beat the simple yardsticks: a straight line through the data is ahead by 23 points of variance explained on the 64 held-out rows. The model's complexity is not buying anything yet. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had. Of the 64 held-out rows where the two differ, the model is closer on 28% (95% interval 19% to 40%). The interval of the margin includes zero: on this many held-out rows the two are not told apart.

Best yardstick's score
0.5175 (95% CI 0.1245 to 0.748)
Best baseline R²
0.5175
Effect size d
-0.5062 (95% CI -0.8499 to -0.2454)
Effect size d magnitude
medium
Margin over the best yardstick
-0.2274 (95% CI -0.4783 to 0.04712)
Margin's interval stays above zero
No
Model MAE
23372.2 (95% CI 19090.4 to 28136.3)
Model
0.2901 (95% CI 0.07231 to 0.3964)
Model R²
0.2901 (95% CI 0.07231 to 0.3964)
Held-out rows
64
Held-out rows without an answer
0
Target
Rprice
Rows where the model is closer
28.13% (95% CI 18.59% to 40.13%)
Rows compared
64
Rows level
0
Regression yardstick trained with gaps filled (too few complete rows)
No
Training rows of the regression yardstick
257

The model explains held-out rows about as well as training rows (held-out R² 0.29 on 64 rows, gap between training and held-out R² 0.177): it generalizes. The structure the formula encodes carries from the training rows to the held-out rows.

Generalization gap
0.1766
N test
64
Target
Rprice
Test error
23372.2 (95% CI 19090.4 to 28136.3)
Test R²
0.2901 (95% CI 0.07231 to 0.3964)
Train error
0.4427
Train R²
0.4667
Verdict
Generalizing

S.3 No range of its inputs where it is markedly weaker

Where it is weak. The largest miss on a held-out row is 99,800 (recorded 207,000, predicted 108,000). 1.6% of the held-out rows are unlike the rows it learned from.

Segment y81ldist in the third quarter (0 to 9.74) performs in line with the aggregate (51 held-out rows): segment R² 0.265 against aggregate R² 0.29 (gap -0.0253). No quality drift detected.

Table 6.2.1 · Rprice by Input Range
SegmentRowsR²Gap to all rowsVerdict
Y81ldist = q3510.2648-0.0253In line with all rows
Rooms = q4330.1717-0.1183In line with all rows
Y81nrinc = false530.3570+0.0669In line with all rows
Year = q3360.2182-0.0719In line with all rows
Y81 = false360.2182-0.0719In line with all rows
Nearinc = false400.3570+0.0669In line with all rows
Y81nrinc = true110.0643-0.2258Too few rows to judge
Cbd = q2100.0696-0.2205Too few rows to judge
Lrprice = q412-2.0190-2.3091Too few rows to judge
Wind = q111-23.0599-23.3500Too few rows to judge

The first 10 rows of 24. The other 14 are in the notebook.

Aggregate MAE
23372.2
Aggregate R²
0.2901
Measure
r2
Target
Rprice

The worst held-out row was missed by 99,810 (target's scale). The 10 rows with the largest absolute residuals out of 64 held-out rows are shown. Each card breaks the failure down into per-term contributions and input values, so the failure mode is inspectable. A large residual with one concentrated term contribution means one term got this row wrong; a large residual with balanced contributions means a structural mismatch. 1 of these 10 rows has at least one input outside the training range (marked "outside training range"): it is an extrapolation, where the model has no anchored evidence; a large residual there reflects missing evidence rather than a failure of the model.

Table 5.1.13 · Worst Predictions
ActualPredictedResidualLandLarea
207,373107,563+99,81043,5608.112
76,805129,184-52,37947,9168.544
134,40982,114+52,29529,1307.519
153,610104,612+48,99764,7457.812
148,000100,453+47,54743,5607.931
31,00075,547-44,54745856.83
36,00080,179-44,17996657.636
52,22795,526-43,29945,7387.783
31,49073,907-42,41782766.842
40,00081,594-41,59483707.868
Additive
Yes
Expected coverage
0.8723
N extrapolated
1
Held-out rows
64
Held-out rows for the interval
64
Rows inside the interval
59
Residual P5
-43166.7
Residual P95
46600.2
Residual std
28555

1 of 64 held-out rows (1.6%) lie further from the training rows than 99 in 100 rows like them would, measured on the 2 inputs the model reads (land, larea). That is a small minority; the held-out figures describe the model on rows like the ones it was trained on.

Table 6.3 · Rows unlike the ones the model learned from
TargetInputs measuredLineHeld-out rowsBeyond the lineShare
RpriceLand, Larea3.0356411.56%
Table 6.3 · Score bins · 24 of 30 rows shown
CountFromTo
200.121
30.1210.243
20.2430.364
40.3640.486
50.4860.607
60.6070.728
50.7280.85
30.850.971
50.9711.093
51.0931.214

The first 10 rows of 24. The other 14 are in the notebook.

S.4 Usable with care: 3 points to watch

What to watch. First: Held-out residuals carry a systematic offset: the model over-predicts by 8,106 on average.

Table 3 · Concerns
SeveritySectionConcernDetail
WarningModelHeld-out residuals carry a systematic offset: the model over-predicts by 8,106 on averageAcross 64 held-out rows, predictions sit 8,106 above the actual values on average, against a residual standard deviation of 28,555: the offset is 28% of the residual spread. The offset is small relative to the residual spread but consistent in direction, so individual predictions lean the same way. Account for it wherever predictions are added up or compared with a threshold, and check on the next run whether it persists.
WarningModelResidual variance grows with predicted value: model accuracy degrades on higher-magnitude predictionsAcross 64 held-out rows split into thirds by predicted value, the residual standard deviation in the top third is 35,436, against 15,318 in the bottom third, 2.31 times as large. The bottom third holds predictions up to 78,031 and the top third predictions from 91,424. Relative to the predictions, the typical miss is 21% of the value in the bottom third and 35% in the top third, 1.65 times as large. The residual spread is 2.31 times as large at the high end: noticeable, not severe. Predictions at the high end deserve caution; a log transform of the target may tighten them.
WarningModelThe model does not beat a straight line through the data on RpriceThe model explains 29% of the variation on the held-out rows, where a straight line through the data explains 52%: 23 points less. The interval of the margin includes zero: on this many held-out rows the two are not told apart. The formula adds little over the reference predictor; whether it is worth its complexity depends on the use. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had.
InfoModel3 columns track Rprice almost completely3 columns each tell almost everything about Rprice on their own, 1 of them as exactly as a copy would. The closest are Lrprice (Spearman's rho = 1 (Pearson r = 0.959)), Lprice (Spearman's rho = 0.955 (Pearson r = 0.921)) and Price (Spearman's rho = 0.955 (Pearson r = 0.952)). The coefficients read the same for a strong input and for a leak. A column that is known before Rprice exists is a strong input, and the quality stated in this report is real. One that was computed from Rprice or recorded after it makes a formula that will not hold on new data. The model reads none of them, so its predictions do not depend on the answer. Each column needs its own answer: confirm it as known before the target, or leave it out and run the analysis again.

The analysis ran with the configuration below. The checks and the verdict of this chapter are read against it.

  • Targets: 1 target: rprice
  • Rows: 321 rows
  • Per-Segment Predictions: not produced, because no segment column was chosen
  • Fairness: compared across 3 categorical columns (nearinc, y81, y81nrinc); which of them are protected attributes is not decided here
  • Out-of-Distribution: held-out rows are scored against the training rows; predictions made outside this report are not scored
  • Provenance & Audit: recorded in the chapter of that name

Across 6 checks (3 pass, 3 warn, 0 fail), the model passes the hard gates but at least one check warns: Usable with care. A limited audience and monitoring of the warning signals come before a broad rollout.

Table 6.6 · Deployment Readiness
NameStatusDetail
GeneralizationPassquality on the held-out rows matches quality on the rows the model was found on
Prediction Interval CoverageWarntoo few held-out rows to judge coverage, largest gap 5.00 percentage points
Baseline Lift for RpriceWarnis not told apart from the best baseline, R² lift -0.227
Model ConstraintsPassevery held-out prediction is finite and within the training range
FairnessWarn3 columns compared, 3 flagged: Nearinc (a group is systematically over- or under-predicted), Y81 (a group is systematically over- or under-predicted), Y81nrinc (a group is systematically over- or under-predicted)
Out-of-DistributionPass1.6% of held-out rows are out of distribution
N critical concerns
0
N fail
0
N pass
3
N warn
3
Overall verdict
Caution

Column Nearinc splits the held-out rows into 2 groups. The prediction-parity ratio is 0.95 (smallest over largest mean prediction), the largest absolute group residual mean is 21,376 (0.749 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. ldist goes closely with nearinc (association 0.84 on a scale from 0 to 1, and 4 more columns at 0.5 or above): a model that reads ldist can tell the groups of nearinc apart, so leaving nearinc out of the inputs does not keep it out of the predictions.

Aggregate residual std
28555
N columns
3
N proxies
10
Proxy threshold
0.5
Table 6.4.1 · Columns
Bias max abs residual meanColumnGroupsInterpretationN groupsN proxiesPrediction parity ratioVerdict
21375.7Nearincfalse, trueColumn Nearinc splits the held-out rows into 2 groups. The prediction-parity ratio is 0.95 (smallest over largest mean prediction), the largest absolute group residual mean is 21,376 (0.749 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. ldist goes closely with nearinc (association 0.84 on a scale from 0 to 1, and 4 more columns at 0.5 or above): a model that reads ldist can tell the groups of nearinc apart, so leaving nearinc out of the inputs does not keep it out of the predictions.250.95Residual bias
11217.7Y81true, falseColumn Y81 splits the held-out rows into 2 groups. The prediction-parity ratio is 0.976 (smallest over largest mean prediction), the largest absolute group residual mean is 11,218 (0.393 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. year goes closely with y81 (association 1.00 on a scale from 0 to 1, and 3 more columns at 0.5 or above): a model that reads year can tell the groups of y81 apart, so leaving y81 out of the inputs does not keep it out of the predictions.240.9758Residual bias
22802.7Y81nrincfalse, trueColumn Y81nrinc splits the held-out rows into 2 groups. The prediction-parity ratio is 0.992 (smallest over largest mean prediction), the largest absolute group residual mean is 22,803 (0.799 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. nearinc goes closely with y81nrinc (association 0.576 on a scale from 0 to 1): a model that reads nearinc can tell the groups of y81nrinc apart, so leaving y81nrinc out of the inputs does not keep it out of the predictions.210.9922Residual bias

S.5 A larger held-out set would firm up the figures

What to collect next. The figures rest on 64 held-out rows.

  • Add data recording the group each row belongs to for the column Nearinc. Reason: the model's errors differ between groups in Nearinc.
  • Add data recording the group each row belongs to for the column Y81. Reason: the model's errors differ between groups in Y81.
  • Add data recording the group each row belongs to for the column Y81nrinc. Reason: the model's errors differ between groups in Y81nrinc.
  • Add more columns or features that may track Rprice more closely. Reason: the model explains 29% of the variation in Rprice and does not beat a straight line through the data.
  • Add columns that help identify or address rows where the predicted value is higher, and record values that may help explain high-magnitude predictions. Reason: model accuracy degrades on higher-magnitude predictions and the residual variance grows with predicted value.

S.6 It shows how Rprice moves with its inputs in these rows, not what causes it

What this model does not show. It was built on Larea from 6.6 to 8.54 and Land from 1710 to 545,000; outside those ranges it has not been checked. 75 slices of the data had too few held-out rows to judge.

Diafunc found a formula that captures the structure in the training data. That structure is correlative: it states what predicts what in this data. It does not establish that any input causes the target.

A correlative model is the right tool for predicting the target on new rows like these, for exploring which inputs carry information about the target, and for an early-stage diagnostic or screening tool.

A correlative model is not the right tool for claiming that intervening on an input will change the target, for attributing a target's values to specific causes for accountability, or for choosing between policies that act on the inputs.

The Quality by Input Range, Fairness and Concerns chapters show the most common confounding patterns in the data.

The model is correlative: it captures which columns predict the target in the training data, not whether changing one would change the other. Predictions are reliable for inputs that look like the training distribution; they are not evidence that intervening on an input will change the target. The 4 upgrade pathways below show what evidence would support a causal claim.

Table 6.5 · Upgrade pathways
NameSummary
Controlled experimentRandomize the input to be acted on, then analyze again. Randomization breaks confounding, so any change in the target is attributable to the input.
Natural experimentFind a setting where the input varied for reasons unrelated to the target, an instrumental variable. Variation of that kind supports causal inference.
Causal graphWrite out what plausibly causes what, then check whether the model's terms are consistent with the conditional independences the graph implies.
Control for confoundersIf a third variable plausibly moves both the input and the target, stratify by it or condition on it before training again. Gaps in the Quality by Input Range chapter are a clue.

Evidence

Every chapter of the analysis in full: the figures the story rests on, and everything else the analysis measured.

1 Headline

The model explains less than half of Rprice from 24 columns, using just 2. It explains 29% of Rprice and relies most on Larea, with which the model's output rises.

  • One target was modeled; 64 rows were held out for checking.
  • A straight line through the data already explained 52%; the model did not beat this baseline, trailing by 23 points of variance explained.
  • The model’s errors differ between groups in Nearinc, Y81, and Y81nrinc.
Best R²
0.2901
Best quality
16.54%
Best target
rprice
Targets
1

2 Stakeholder Summary

The model predicts Rprice and relies most on Larea, which accounts for 76% of its reliance. It explains 29% of the variation in Rprice, with the strongest baseline, a straight line through the data, explaining 52%. This means the model did not beat the baseline, finishing 23 points behind on variance explained. Raising Larea from 7.30 to 7.81 raises the model’s output by about 18,918 on average, with everything else as observed.

Because the model did not outperform a straight line through the data, individual predictions are not more useful than that yardstick. The model uses specialist functions in a formula of moderate length, and its errors differ between groups for Nearinc, Y81, and Y81nrinc. The next step is to validate the core relationships or seek simpler or stronger formulas before relying on the model for decision-making.

3 Concerns

Table 3 · Concerns
SeveritySectionConcernDetail
WarningModelHeld-out residuals carry a systematic offset: the model over-predicts by 8,106 on averageAcross 64 held-out rows, predictions sit 8,106 above the actual values on average, against a residual standard deviation of 28,555: the offset is 28% of the residual spread. The offset is small relative to the residual spread but consistent in direction, so individual predictions lean the same way. Account for it wherever predictions are added up or compared with a threshold, and check on the next run whether it persists.
WarningModelResidual variance grows with predicted value: model accuracy degrades on higher-magnitude predictionsAcross 64 held-out rows split into thirds by predicted value, the residual standard deviation in the top third is 35,436, against 15,318 in the bottom third, 2.31 times as large. The bottom third holds predictions up to 78,031 and the top third predictions from 91,424. Relative to the predictions, the typical miss is 21% of the value in the bottom third and 35% in the top third, 1.65 times as large. The residual spread is 2.31 times as large at the high end: noticeable, not severe. Predictions at the high end deserve caution; a log transform of the target may tighten them.
WarningModelThe model does not beat a straight line through the data on RpriceThe model explains 29% of the variation on the held-out rows, where a straight line through the data explains 52%: 23 points less. The interval of the margin includes zero: on this many held-out rows the two are not told apart. The formula adds little over the reference predictor; whether it is worth its complexity depends on the use. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had.
InfoModel3 columns track Rprice almost completely3 columns each tell almost everything about Rprice on their own, 1 of them as exactly as a copy would. The closest are Lrprice (Spearman's rho = 1 (Pearson r = 0.959)), Lprice (Spearman's rho = 0.955 (Pearson r = 0.921)) and Price (Spearman's rho = 0.955 (Pearson r = 0.952)). The coefficients read the same for a strong input and for a leak. A column that is known before Rprice exists is a strong input, and the quality stated in this report is real. One that was computed from Rprice or recorded after it makes a formula that will not hold on new data. The model reads none of them, so its predictions do not depend on the answer. Each column needs its own answer: confirm it as known before the target, or leave it out and run the analysis again.

How to read this

Data-quality, model-quality, and deployment-risk issues detected during the analysis appear below, ordered by severity. Critical concerns are listed first; Warning and Info entries follow.

4 Discovered Relations

Table 4 · Quality and error measured on the 64 held-out rows
TargetFormulaQualityComplexityError
RpriceRprice = 77700 − 0.9 × lerp(Larea, 7.35, Land)16.54%8.9%70.99%

How to read this

One model per target, each a formula with its quality, complexity and error. Higher quality at lower complexity describes the data most clearly. The findings below read the models together; the per-target chapters in the Model section carry each formula, its predictions and its diagnostics.

5 Model

The model relies on Larea more than other inputs, and its ability to generalize drops on held-out rows compared to training rows. Several warnings were flagged: see the Concerns chapter for details.

Most notable in the Model section:

  • Larea carries most of the model (76% of its reliance on inputs)
  • The model explains 18 points less variance on held-out rows than on training rows (generalizing)
  • Held-out errors lean one way: the model over-predicts by 8,106 on average, about 0.284 standard deviations

How to read this

The Model section presents one model per target: its formula, its predictions on held-out rows, the inputs it relies on, the baselines it is measured against, and diagnostics of how it behaves. With more than one target, the Composite Model, Multi-Output and Leaderboard chapters read the models as a set.

5.1 Rprice

The Model section builds one model for Rprice: a formula that expresses it through the other columns. Diafunc searches the space of mathematical expressions for the formula that balances accuracy against simplicity, so the result is not a black box but an expression that can be read, evaluated and questioned. The chapters that follow show the formula, score it on rows it never saw during training, and take it apart term by term.

5.1.1 Rprice Model

The model's output relies on Larea, with the strongest input importance at 76%, and rises with Larea. It also tracks Land, where the relationship depends on the range.

Rprice = 77700 − 0.9 × lerp(Larea, 7.35, Land)f:(Land,Larea)↦RpriceLand,Larea,Rprice∈RRprice=77700−0.9×lerpLarea,7.35,Land
Rprice = 77700 − 0.9 × lerp(Larea, 7.35, Land)
Quality 16.54% · Error 70.99% on 64 rows the search never saw · Found on 257 rows · Complexity 8.9%
Table 5.1.1 · Constants of the formula
ValueRoleInputUnit
77,700InterceptUnit of the model's output
0.9Coefficient
7.35Argument
Rows the search trained on, measured
130
Rows the search trained on
130
Held out R²
0.2901 (95% CI 0.07231 to 0.3964)
Constant base
77,700
Constants in the formula
3
Terms that read an input, constants multiplied in
1
Training rows
257
Held-out rows
64
Inputs the formula reads
2
Terms in the formula
2
Inputs the formula does not read
Age, Agesq, Area, Baths, Cbd, Dist, Intst, Ldist, Lintst, Lintstsq, Lland, Nbh, Nearinc, Rooms, Wind, Y81, Y81ldist, Y81nrinc, Year

How to read this

The three gauges score the model on the held-out rows: rows that were set aside before the search and that the model never saw. They are the figures to quote for how the model does on new data.

Table 5.1.1 · Rprice Model
MetricMeaning
QualityHow well the model predicts on the held-out rows, marked down for the size of its formula: 100% is an exact prediction, 0% is no better than always predicting the average. It is not R², and sits at or just below it
ErrorHow far the predictions lie from the held-out values, as a share of the miss of always predicting the average: 0% is an exact prediction, 100% is no better than the average
ComplexityThe size of the formula as a share of the complexity the search was allowed: a low figure is a short formula

The figures under During the search are the same measures on the training rows. They are what the search ranked its candidates by, so they describe the search and tend to flatter the model it chose. The Generalization chapter sets the training rows beside the held-out rows, and the Baselines chapter sets the model beside its baselines. The time and credit gauges account for the computation that produced the model, not for the model itself.

5.1.2 Prediction

The model predicts Rprice for any row that supplies the other columns. This chapter applies it to the held-out rows, the rows set aside before training, so that the predictions can be compared with values the model never saw. The rows themselves and the function that predicts them are kept as reference under Foundation, with the processed data.

5.1.2.1 Prediction Chart

R² of 0.29 on 64 rows the search never saw (95% interval 0.0723 to 0.396): weak. The model captures only a small part of the variance; either the relevant inputs are missing, or the target depends on interactions the formula did not recover. Mean absolute error is 23,372 on the target's scale.

Error
70.99% (95% CI 60.36% to 92.77%)
MAE
23372.2 (95% CI 19090.4 to 28136.3)
N
64
Training rows
257
Quality
16.54% (95% CI 0% to 29.04%)
R²
0.2901 (95% CI 0.07231 to 0.3964)
Rows found on
257
Target
Rprice

How to read this

Actual Rprice on the horizontal axis against the model's predicted value on the vertical axis, for the held-out rows the model never saw during training. Points on the diagonal are perfect predictions; systematic vertical offsets reveal bias, vertical spread reveals noise, and points far from the diagonal are rows the model handles poorly. With 10,000 held-out rows or more, the rows are counted instead of drawn one point each: each cell is a range of the actual and a range of the predicted value, its color the number of rows that fall in both, and the cells whose two ranges match hold the rows the model predicts well.

5.1.2.2 Residual Distribution
Bias
-8105.8 (95% CI -14683 to -1239.8)
N
64
P50
-8577.1 (95% CI -20003.7 to -4698.2)
P90
28347.6 (95% CI 7368.1 to 47546.6)
P95
46600.2 (95% CI 21629.4 to 52295)
P99
69875.5 (95% CI 36774.9 to 99809.9)
Std
28555 (95% CI 21629.2 to 35230.8)

How to read this

How the held-out residuals (actual Rprice minus predicted) are distributed. Bias is their mean: a value away from zero means the model is systematically high or low rather than merely imprecise. The standard deviation and the percentiles describe the spread, and the tail percentiles say how large the worst misses get. Every figure carries a bootstrap 95% confidence interval; with a small held-out set those intervals are wide, and that width is itself the finding.

5.1.3 Which inputs matter most

Measured on 25 of the 257 rows the model was found on, which the search set apart for it. On so few rows the order of the inputs is indicative, not settled. The model's output rises with Larea on every one of the 8,320 steps between observed values. With Land the model's output rises on 5,824 and falls on 2,496 of the 8,320 steps between observed values. Moving Larea from 7.30 to 7.81, its first and third quartile, with every other input as observed raises the model's output by 18,918 on average (by between 788 and 251,126 on single rows). These are measurements on the rows the search trained on, inside the observed ranges; outside them nothing is claimed.

Table 5.1.3 · Which inputs matter most
InputShare of importancePermutation importanceSensitivityResidual correlationOccurrencesModel's output as the input risesChange, first to third quartile
Larea75.76%0.883736333.7-0.20621Rises (8320 of 8320 steps)+18918.2 (7.297 to 7.81)
Land24.24%0.28270.39590.19671Rises and falls (rises on 5824, falls on 2496 of 8320 steps)+5277.8 (16,935 to 45,973)
Rows the search trained on, measured
130
N rows
25
Inputs the formula does not read
Age, Agesq, Area, Baths, Cbd, Dist, Intst, Ldist, Lintst, Lintstsq, Lland, Nbh, Nearinc, Rooms, Wind, Y81, Y81ldist, Y81nrinc, Year

How to read this

The chart ranks the inputs of the Rprice model by permutation importance: how much the model's error rises when that input's values are shuffled across rows so that it carries no information. A longer bar means the model relies on the input more. The order says which columns the model reads most, and those are the columns where measurement quality and data collection matter most. A short bar marks an input the model barely uses, which is a fact about this model and not a verdict on the column. Importance describes the model's reliance and nothing more: an input can rank high because it stands in for something the data does not contain. Where the search measured it, an input also carries the direction in which the model's output moves as the input rises, counted on the rows the search trained on, and the change of the output when the input alone goes from its first to its third quartile; both describe the formula inside the observed range and say nothing about causes.

5.1.4 Baselines

The model does not beat the simple yardsticks: a straight line through the data is ahead by 23 points of variance explained on the 64 held-out rows. The model's complexity is not buying anything yet. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had. Of the 64 held-out rows where the two differ, the model is closer on 28% (95% interval 19% to 40%). The interval of the margin includes zero: on this many held-out rows the two are not told apart.

Best yardstick's score
0.5175 (95% CI 0.1245 to 0.748)
Best baseline R²
0.5175
Effect size d
-0.5062 (95% CI -0.8499 to -0.2454)
Effect size d magnitude
medium
Margin over the best yardstick
-0.2274 (95% CI -0.4783 to 0.04712)
Margin's interval stays above zero
No
Model MAE
23372.2 (95% CI 19090.4 to 28136.3)
Model
0.2901 (95% CI 0.07231 to 0.3964)
Model R²
0.2901 (95% CI 0.07231 to 0.3964)
Held-out rows
64
Held-out rows without an answer
0
Target
Rprice
Rows where the model is closer
28.13% (95% CI 18.59% to 40.13%)
Rows compared
64
Rows level
0
Regression yardstick trained with gaps filled (too few complete rows)
No
Training rows of the regression yardstick
257

How to read this

The table scores the model for Rprice against simple yardsticks on the same held-out rows. A yardstick is a predictor simple enough to need no explanation: a constant (the average, the middle value, the most common class or level, the average count), a plain regression on the numeric inputs (a straight line, a logistic regression, a Poisson regression), and for a series over time a forecast from its own past (the last value, the value one season back, the recent trend extended). Which of them appear depends on what kind of column Rprice is. They are yardsticks only: the analysis has one model, and it is the last row. A regression yardstick answers every row the model answers: where one of its numeric inputs is empty, it reads the average of that input over the training rows. Nothing is filled in for the model. A yardstick whose row is empty could not be scored, and the summary then says what the margin is measured against.

Read the columns from top to bottom. Each kind of target is scored by the measure it is judged on: R² and MAE for a continuous value, with MASE added for a series (below 1 means a typical miss is smaller than the series' usual step from one row to the next); accuracy and balanced accuracy for a yes/no value, with the Brier score where a probability is predicted; accuracy and macro F1 for a class; weighted kappa and the distance in levels for an ordered level; Poisson deviance and RMSE for a count. When the model clears every yardstick by a wide margin, its structure earns its place. When it is level with the regression yardstick, a plain regression describes the data as well. When it is level with the constant, the inputs carry no usable signal for Rprice.

The margin over the best yardstick comes with an interval. When that interval includes zero, the lead is not settled on this many held-out rows. The share of rows on which the model is closer than the best yardstick is counted only on the rows where the two differ; for a yes/no value or a class these are the rows where the two disagree.

5.1.5 Generalization

The model explains held-out rows about as well as training rows (held-out R² 0.29 on 64 rows, gap between training and held-out R² 0.177): it generalizes. The structure the formula encodes carries from the training rows to the held-out rows.

Generalization gap
0.1766
N test
64
Target
Rprice
Test error
23372.2 (95% CI 19090.4 to 28136.3)
Test R²
0.2901 (95% CI 0.07231 to 0.3964)
Train error
0.4427
Train R²
0.4667
Verdict
Generalizing

How to read this

The chart sets the score of the Rprice model on the rows it trained on beside its score on the held-out rows it never saw, and the gap between the two bars is the finding. The score is the one that suits the target: R² for a quantity, balanced accuracy for a yes/no target, weighted kappa for ordered levels. The held-out figure carries a bootstrap 95% confidence interval, which says how far it could move with a different held-out draw; the training bar has none, because it is measured on the rows that shaped the model.

The verdict reads from the two figures. Generalizing: the held-out score is strong, or training and held-out agree closely; the structure the formula encodes carries to new rows. Overfit: training is far ahead of held-out and the held-out score is not strong on its own; the formula has absorbed noise from the training rows, and the training figure overstates what to expect on new data. Underfit or noisy: held-out is far ahead of training; this is not strength, it usually means an outlier or a flat region in the training rows that the held-out rows happen to avoid. Insufficient: fewer than 30 held-out rows and no strong held-out score, too little evidence for a negative verdict. A model that predicts the same value for every held-out row gets no verdict: there is nothing that could carry over. In every case the width of the held-out interval says how much weight the gap can bear.

5.1.6 How each term responds to its inputs

The formula decomposes into 2 additive terms: clean separable structure. Each curve below shows one term's response to its free inputs, with the other inputs held at the training median. Steep curves carry the prediction; flat curves are decorative.

Additive
Yes
Rows the search trained on, measured
130
Terms in the formula
2
Notation
77700 − 0.9 × lerp(Larea, 7.35, Land)
Target
Rprice

How to read this

The formula for Rprice is split into its additive terms, and each term is traced as one of its inputs moves across the training range while the other inputs stay at their medians. Each curve is therefore a profile of the term at the typical row, not an average over the data: it shows what the term contributes at every value of that input when everything else is ordinary.

A term that depends on several inputs gets one curve per input; a constant term gets none. Terms the formula subtracts are drawn with their sign, so a curve below zero pulls the prediction down. At any fixed row, the terms add up to the model's prediction. Where the search measured it, a term also carries its share of the variation of the model's output on the rows the search trained on; that is a share of what the formula returns, not of Rprice itself. The notebook lets the reader trim each range to the central 95% of the input. When the formula has no additive structure, the chapter traces the whole formula as a single term along each of its inputs, and the Sensitivity chapter is the better read for how the inputs combine.

5.1.7 How the formula behaves outside the data

How the model behaves beyond the observed range of larea, land could not be determined from the formula, so extrapolation there is not assessed. Extrapolate with caution.

Bounded at infinity
No
Rows the search trained on, measured
130
Inputs the formula reads
2
N singularities
0
Target
Rprice
Table 5.1.7 · Limits at neg infinity
InputKind
LareaUnknown
LandUnknown
Table 5.1.7 · Limits at pos infinity
InputKind
LareaUnknown
LandUnknown

How to read this

The formula for Rprice is read as a mathematical object, without data, to state how it behaves beyond the range it was trained on. A black box cannot answer that question; a formula can. For each input the chapter reports the limit as the input grows without bound in either direction (whether the formula diverges, settles to a finite value or oscillates, with the sign tracked exactly, so that a term like minus two times the cube of the input is reported as diverging downward on the right and upward on the left), whether the formula stays finite in every direction, and any singularity with its solved location: a division whose denominator reaches zero, a logarithm or square root whose argument leaves its domain, or a periodic pole.

The inputs along which the formula diverges or meets a singularity are listed as the risky ones; along them the formula's shape gives no support for extrapolation. A formula that stays finite everywhere with no singularity carries no structural hazard when extrapolated; the data still ends where it ends, and a prediction outside the training range remains unsupported by evidence. The plain-language summary at the end states the same facts in one paragraph.

5.1.8 Contributions per term
Additive
Yes
N rows
5

The notebook draws this chapter's figure from each held-out row; the row-by-row detail is there.

How to read this

For up to five representative held-out rows, the chapter draws one bar chart showing how each additive term of the formula builds that row's prediction of Rprice. The five rows are the worst prediction (largest absolute residual), the best (smallest), the median miss, the row with the smallest prediction and the row with the largest. When two roles fall on the same row, which is common with few held-out rows, the earlier role keeps it and fewer than five charts appear.

In each chart the bars stack the terms' contributions at the row's input values, terms the formula subtracts sit below zero, and the bars sum to the prediction. The dashed line marks the actual value; the distance between it and the top of the stack is the residual. The bars are evaluated at the same precision as the predictions, so they agree with the Prediction Chart. When the formula has no additive structure, each chart holds one bar for the whole formula, and the Sensitivity chapter is the better read for how individual inputs move it.

5.1.9 What each term is worth

Full-formula R² is 0.29: weak overall quality. Ablation deltas are not very informative when the baseline quality is low; large drops are best-effort hints rather than definitive structural claims.

Additive
Yes
Baseline R²
0.2901
Held-out rows
64

The notebook draws this chapter's figure from each held-out row; the row-by-row detail is there.

How to read this

For each additive term of the formula for Rprice, the chart shows how much the held-out R² drops when that term is replaced by its mean contribution, so that it no longer varies from row to row. The bars are sorted from the largest drop down, and the dashed line marks the R² of the full model.

A large drop means the term carries real predictive work; the complexity it adds is paid for. A small drop means the term adds little beyond its average value: it may refine a stronger term, or its curve in the chapter "How each term responds to its inputs" may be nearly flat. A negative drop means the held-out R² rises without the term, which is a sign that the term tracks noise in the training rows. Each figure describes what a term contributes inside the model as discovered; the chart diagnoses the model, it is not a guide to removing terms. When the formula has no additive structure, the chart shows one bar for the whole formula.

5.1.10 What-If
Table 5.1.10 · The prediction at the typical row: every input at its median or most common value
ItemValue
Land43,560
Larea7.652
Term 1 adds77,700
Term 2 adds11816.9
Prediction89516.9
Additive
Yes
Expected coverage
0.8723
Held-out rows for the interval
64
Rows inside the interval
59
Terms in the formula
2
Residual P5
-43166.7
Residual P95
46600.2
Residual std
28555

How to read this

The chapter shows the model's prediction for Rprice at one input combination, with the contribution of each term stacked beneath it. The inputs start at their training medians, so the first view is the prediction for a typical row, and each input can move within the range the training rows cover; the notebook lets the reader change the inputs and watch the prediction and the stack update.

In the stack, a long bar marks the term that dominates at this input combination, often the one whose removal costs most in the chapter "What each term is worth". A short bar marks a term that makes a small adjustment here; the chapter "How each term responds to its inputs" shows whether it stays small everywhere or grows toward the edges of the range. A negative bar marks a term that pulls the prediction down at this point, either because the formula subtracts it or because it evaluates below zero.

The inputs stay inside the training range for a reason: the formula computes a value anywhere, but outside that range no training row supports it. When the formula has no additive structure, the stack holds a single bar for the whole formula and the inputs move the prediction directly.

5.1.11 Sensitivity

The prediction responds to 2 inputs. The formula is additive: each curve below shows the change in the prediction per unit of one input (∂ŷ/∂xᵢ) independently. Flat curves mean a constant linear response, bent curves a non-linear response that depends on the position in the range, and zero crossings a response that reverses sign between regions.

Additive
Yes
Terms in the formula
2

How to read this

For each input, the chart traces the rate at which the predicted Rprice changes as that input moves across its training range, with the other inputs held at their medians. The rate is the local slope of the prediction with respect to the input, measured by a small finite difference at each point. A table summarizes each input with the mean, minimum and maximum of the absolute slope, which ranks the inputs by how strongly the prediction responds to them.

A flat curve means the model is close to linear in that input: the prediction responds at the same rate everywhere in the observed range, which is easy to state and says nothing about values beyond that range. A curve that grows or shrinks marks diminishing or accelerating returns, and whether that is plausible is a question for the domain. A curve that crosses zero means the prediction rises with the input in one part of the range and falls in another, which is either a real optimum or a sign that the formula bends around a few rows. A curve with a large magnitude marks an input to which the prediction is highly responsive; the chapter "What each term is worth" says whether the term behind it also carries predictive work.

In the table, the mean absolute slope ranks the inputs by typical responsiveness. A minimum close to the maximum means the responsiveness is nearly constant; a wide gap means it varies strongly across the range. The chapter reads the whole formula, so it applies whether or not the formula splits into terms.

5.1.12 How two inputs act together

The formula couples land and larea: on 130 of 130 row pairs the mixed difference was not zero, so in the model's output the two act together (measured, not assumed). This describes the formula; it does not say that the data hold no other interaction. The heatmaps below show how the prediction's response to one input depends on another (the others held at the median).

Table 5.1.12 · Input pairs tested inside the formula
InputsIn the model's outputRow pairsPairs with a differenceMean absolute mixed difference
Land, LareaAct together13013011562.9
Table 5.1.12 · The prediction across Land (columns) and Larea (rows), the other inputs at their typical values
Larea \ Land1710100,399199,088297,777396,466495,155
6.6276570.711738.5-53093.6-117,926-182,758-247,590
6.96877105.643160.29214.8-24730.6-58,676-92621.4
7.31677640.474581.871523.268464.665,40662347.4
7.66378175.3106,003133,832161,660189,488217,316
8.01178710.2137,425196,140254,855313,570372,285
8.35979,245168,847258,448348,050437,652527,254
Additive
Yes
Inputs in different terms act separately
Yes
Rows the search trained on, measured
130
Terms in the formula
2
Input pairs not tested
0

How to read this

When a single term of the formula combines two or more inputs, the prediction's response to one of them depends on the value of the other. This chapter draws one surface for each such pair: the predicted Rprice across the plane of the two inputs, with every other input held at its training median. Pairs that never share a term have no interaction by construction, and the curves in the chapter "How each term responds to its inputs" already say everything about them. Where the search tested a pair, the summary says on how many pairs of rows the two inputs acted together in the model's output; that is a statement about the formula, not a finding that the data hold no other interaction.

Parallel stripes mean the two inputs act almost independently, so the surface is close to a sum of two one-input curves. Curved contours mean a real interaction, where the slope along one input changes with the other; the location of the peak or trough matters. A saddle, rising along one axis and falling along the other, marks a trade-off between the two. A sharp edge or cliff means the formula encodes a change of regime, and the curve for the same input in the chapter "How each term responds to its inputs" shows whether the bend appears there too.

The grid is deliberately coarse, and the color scale runs from low to high predictions. When the formula has no additive structure, the single whole-formula term holds every input, so every pair appears; the number of surfaces is capped for models with many inputs.

5.1.13 Worst Predictions

The worst held-out row was missed by 99,810 (target's scale). The 10 rows with the largest absolute residuals out of 64 held-out rows are shown. Each card breaks the failure down into per-term contributions and input values, so the failure mode is inspectable. A large residual with one concentrated term contribution means one term got this row wrong; a large residual with balanced contributions means a structural mismatch. 1 of these 10 rows has at least one input outside the training range (marked "outside training range"): it is an extrapolation, where the model has no anchored evidence; a large residual there reflects missing evidence rather than a failure of the model.

Table 5.1.13 · Worst Predictions
ActualPredictedResidualLandLarea
207,373107,563+99,81043,5608.112
76,805129,184-52,37947,9168.544
134,40982,114+52,29529,1307.519
153,610104,612+48,99764,7457.812
148,000100,453+47,54743,5607.931
31,00075,547-44,54745856.83
36,00080,179-44,17996657.636
52,22795,526-43,29945,7387.783
31,49073,907-42,41782766.842
40,00081,594-41,59483707.868
Additive
Yes
Expected coverage
0.8723
N extrapolated
1
Held-out rows
64
Held-out rows for the interval
64
Rows inside the interval
59
Residual P5
-43166.7
Residual P95
46600.2
Residual std
28555

How to read this

The held-out rows on which the model misses Rprice by the most, sorted by absolute residual with the largest first and capped at ten. Each card shows the row's actual value, the prediction, the residual, the contribution of every additive term at that row and the row's input values. R² and MAE say how good the model is on average; these cards say where it fails and let each failure be read input by input and term by term.

A large residual with one dominant term means that term produced the wrong value at this row; the row's value for that input, read against its curve in the chapter "How each term responds to its inputs", shows whether it fell in a region the formula models poorly. A large residual with balanced contributions means every term was a little off and the sum still missed, which points to a shape the formula cannot represent in that region. Residuals that change sign from card to card mean the misses are not a consistent bias but scattered errors on particular rows. When the formula has no additive structure, each card shows a single whole-formula contribution.

5.1.14 Prediction Interval Coverage

Coverage is counted on 64 rows, where one row moves it by 2 points. That is too coarse to judge whether the interval is well calibrated; the table shows the counts.

Table 5.1.14 · Prediction Interval Coverage
Nominal levelHalf-widthEmpirical coverage
50%19260.129 of 64 (45.31%)
80%36594.848 of 64 (75%)
90%46968.959 of 64 (92.19%)
95%55966.863 of 64 (98.44%)
Largest miss (percentage points)
5
Held-out rows
64
Rows inside the interval
29, 48, 59, 63
Residual P2.5
-44335.5
Residual P97.5
50398.9
Verdict
insufficient-data

How to read this

Each held-out prediction of Rprice carries a prediction interval, and this chapter checks whether that interval covers the actual value as often as it claims. The interval reported with each prediction is built from the held-out residuals themselves: its bounds are the 2.5th and 97.5th percentiles of the residuals added to the prediction, so it is asymmetric when the residuals are skewed and assumes no particular error distribution.

The coverage table checks a second, simpler band: the symmetric interval that a normal error distribution with the residuals' standard deviation would give. At four nominal levels (50%, 80%, 90% and 95%) it lists the half-width of that normal band and the share of held-out rows whose actual value falls inside it. Empirical coverage close to nominal means the residuals are close enough to normal that the two bands agree and the interval can be taken at face value. Empirical coverage below nominal means the normal band is too narrow, the residuals have heavier tails than a normal distribution, and the percentile-based interval is the one to trust. Empirical coverage above nominal means the normal band is wider than the data needs, usually because a few large residuals inflate the standard deviation. Every figure comes from the same held-out residuals the intervals are built from.

5.1.15 Regression Diagnostics

64 held-out residuals, with a spread up to 99,810 in absolute terms. Mean residual is -8,106, close to zero. The model is approximately unbiased on these held-out rows (a small held-out set makes this less reliable; the Prediction Chart chapter states R² and the mean absolute error with their 95% confidence intervals). The charts below check whether the errors look like noise: a pattern in residuals-vs-fitted or a curving Q-Q means a structural miss. The held-out rows are a random draw, so their order carries no time and the autocorrelation of the residuals is not assessed.

Abs residual vs predicted r
0.2114
Bias
-8105.8
Bias over residual sd
-0.2839
Held-out rows
64
Order
shuffled
Residual sd
28555
Share within normal band
0.9531

The notebook draws this chapter's figure from each held-out row; the row-by-row detail is there.

How to read this

Four residual diagnostics for Rprice, all computed from the held-out residuals. Each asks in its own way whether the model's errors look like noise, which is what remains when a model has captured all the structure the data holds.

Residuals against predicted values: Each held-out row is plotted with the model's prediction on the horizontal axis and its residual on the vertical axis, around a zero line. A symmetric cloud around zero means errors of similar size across the predicted range and no systematic bias. A funnel or fan means the size of the errors depends on the predicted value: the model is more uncertain in some regions than in others, and its prediction interval is not equally trustworthy everywhere. A curved pattern, a smile or a frown, means the model misses a structural component; the chapters "How each term responds to its inputs" and "How two inputs act together" show the shapes the formula does capture.

Normal Q-Q plot: The sorted residuals are plotted against the quantiles of a normal distribution, with the diagonal marking a perfectly normal set of residuals. Points that hug the line mean approximately normal residuals, so the normal band in the Prediction Interval Coverage chapter will hold. An S-shape means heavier or lighter tails than normal; heavier tails (below the line at the left, above it at the right) mean the model occasionally produces large errors. A banana or arc means skewed residuals: the model is systematically high or low in part of the range.

Histogram: The residuals binned into equal-width intervals. A bell centered on zero means symmetric errors with no bias. An off-center bell means the model is biased, with a mean error away from zero. A long tail or two peaks mean outliers or two regimes inside the prediction range.

Autocorrelation: For each lag, the correlation of the residual sequence with itself shifted by that many rows, with the band that pure noise would stay inside at plus or minus 1.96 divided by the square root of the row count. Bars inside the band mean uncorrelated residuals with no order-dependent structure left. A bar at lag 1 outside the band means the residuals of neighboring dates correlate: structure in time that the model has not captured. A wave across the lags means periodic structure remains, and the Time Series chapter in the Data section is where to look for its period.

Autocorrelation carries meaning only when the row order itself does. When the data has a date column, the held-out rows are the most recent ones in date order, and the autocorrelation is assessed. Without a date column the held-out rows are a random draw: the row after a held-out row is an arbitrary other row, so the autocorrelation is not assessed and the summary above says so.

5.1.16 Search Progress

The search ran for 267 generations, ending at quality 0.334 (gain +0.178 from generation 2). The curve shows a substantial climb. The last improvement came after 31% of the search time: the search had settled well before it stopped, so more time alone is unlikely to change the result.

First generation
2
First quality
15.63%
Last generation
268
Last improvement ms
844,507
Last quality
33.43%
Max quality
33.43%
N points
200
N total generations
267
Search time ms
2,732,314
Stop reason
Maximum runtime reached

How to read this

The line traces the best quality the search for the Rprice model had reached at each generation, from the first generation to the last. Quality is the technology's combined score on the training rows, the figure the Model Metrics chapter defines. A curve that flattens well before the last generation means the search converged: the budget was more than enough, and the model is the one the search settled on. A curve still rising at the last generation means the budget ended before the search settled, and the model is the best the search reached within it. The chart describes the search, not the data: it says whether the search had time to settle, not whether more rows would help.

5.1.17 Model Constraints

Across 64 held-out predictions, the formula is mathematically well-behaved: every output is finite, stays inside the training-target ±2σ band, and no prediction runs away beyond the bounded-prediction limit. It is safe to consume downstream at face value.

Table 5.1.17 · Model Constraints
NameStatusDetail
Output coherentPassno non-finite predictions across 64 rows
Within domainPassno rows outside the training target's ±2σ range across 64 rows
Bounded predictionsPassno runaway predictions beyond ±10σ across 64 rows
Held-out rows
64
N non finite
0
N outside domain
0
N runaway
0
Verdict
Clean

How to read this

Sanity checks on the Rprice formula, run against the held-out rows. Three checks: every prediction is a finite number (output coherent), predictions stay within the range of the training target widened by two standard deviations on each side (within domain), and no single prediction runs away more than ten standard deviations from the training mean (bounded predictions).

These are pass or fail guards, not quality measures: a formula can clear all three and still predict poorly. A failure is the stronger signal. It means the formula misbehaves numerically somewhere in the data it was shown, and any downstream use of it should wait until that is understood.

5.1.18 Modeling

Measured during the search, on the rows the model was trained on

MAE of a straight line through the inputs (same rows)
9196.6
RMSE of a straight line through the inputs (same rows)
12360.4
Training MAE (rows the search trained on)
18,399
MAE of always predicting the mean (same rows)
23,769
RMSE of always predicting the mean (same rows)
29331.2
Training R² (rows the search trained on)
0.3766
Training RMSE (rows the search trained on)
23158.8
Rows the search trained on, measured
130
Error
44.27%
Inputs the straight line used
18
N train
257
Quality
33.43%

How to read this

The function below trains the model for Rprice on the processed data. Its output is the model that every other chapter of this section scores and dissects.

5.2 Findings

4 findings from the model diagnostics (3 Warning, 1 Info) are listed in the Concerns chapter at the top of the analysis.

6 Implications

The Implications section states how to act on the models: the Stakeholder Summary and the Recommended Next Data, where the models hold and where they weaken (Quality by Input Range, Rows unlike the ones the model learned from, Per-Segment Predictions, Fairness), what to weigh before relying on them (Causal Honesty, Deployment Readiness), and the Provenance & Audit trail a reviewer needs to retrace the analysis.

6.1 Recommended Next Data

  • Add data recording the group each row belongs to for the column Nearinc. Reason: the model's errors differ between groups in Nearinc.
  • Add data recording the group each row belongs to for the column Y81. Reason: the model's errors differ between groups in Y81.
  • Add data recording the group each row belongs to for the column Y81nrinc. Reason: the model's errors differ between groups in Y81nrinc.
  • Add more columns or features that may track Rprice more closely. Reason: the model explains 29% of the variation in Rprice and does not beat a straight line through the data.
  • Add columns that help identify or address rows where the predicted value is higher, and record values that may help explain high-magnitude predictions. Reason: model accuracy degrades on higher-magnitude predictions and the residual variance grows with predicted value.

6.2 Quality by Input Range

Quality by Input Range splits the held-out rows by every usable column besides the target and compares the model's quality on each segment with its quality on all held-out rows. A column the model does not read, such as one left out because it leaks the target, is split as well: its segments show where the predictions are weak, not what the model uses. For a number the quality is R², the share of variation explained; for a yes-or-no target it is accuracy, the share of rows classified correctly, the measure the Classification Performance chapter reports. A numeric column is split into quartiles; a categorical column with at most ten values is split by value. A segment with fewer than three held-out rows is left out, because its gap would be noise.

The chart is sorted from the worst gap to the best. Each bar is the segment's figure minus the figure on all held-out rows.

  • Negative bars: the model is weaker on these rows than on all held-out rows. A segment 0.25 or more below the overall value (25 points of accuracy) reads much weaker; a segment 0.10 to 0.25 below reads weaker. These are the segments to treat with caution.
  • Positive bars: the model is stronger on these rows than overall. A strong segment can mask weakness elsewhere, so the negative segments come first.
6.2.1 Rprice by Input Range

Segment y81ldist in the third quarter (0 to 9.74) performs in line with the aggregate (51 held-out rows): segment R² 0.265 against aggregate R² 0.29 (gap -0.0253). No quality drift detected.

Table 6.2.1 · Rprice by Input Range
SegmentRowsR²Gap to all rowsVerdict
Y81ldist = q3510.2648-0.0253In line with all rows
Rooms = q4330.1717-0.1183In line with all rows
Y81nrinc = false530.3570+0.0669In line with all rows
Year = q3360.2182-0.0719In line with all rows
Y81 = false360.2182-0.0719In line with all rows
Nearinc = false400.3570+0.0669In line with all rows
Y81nrinc = true110.0643-0.2258Too few rows to judge
Cbd = q2100.0696-0.2205Too few rows to judge
Lrprice = q412-2.0190-2.3091Too few rows to judge
Wind = q111-23.0599-23.3500Too few rows to judge

The first 10 rows of 24. The other 14 are in the notebook.

Aggregate MAE
23372.2
Aggregate R²
0.2901
Measure
r2
Target
Rprice

6.3 Rows unlike the ones the model learned from

1 of 64 held-out rows (1.6%) lie further from the training rows than 99 in 100 rows like them would, measured on the 2 inputs the model reads (land, larea). That is a small minority; the held-out figures describe the model on rows like the ones it was trained on.

Table 6.3 · Rows unlike the ones the model learned from
TargetInputs measuredLineHeld-out rowsBeyond the lineShare
RpriceLand, Larea3.0356411.56%
Table 6.3 · Score bins · 24 of 30 rows shown
CountFromTo
200.121
30.1210.243
20.2430.364
40.3640.486
50.4860.607
60.6070.728
50.7280.85
30.850.971
50.9711.093
51.0931.214

The first 10 rows of 24. The other 14 are in the notebook.

How to read this

Out-of-distribution detection finds held-out rows that lie far from the training rows of a model. It looks only at the numeric inputs the model reads. Each of them is standardized against the mean and standard deviation of the training rows, and the standardized values are combined into one distance per row. The line a row is read against depends on the number of inputs: it is the distance that 99 in 100 rows like the training rows stay below. A row beyond the line is flagged as out of distribution, and the held-out figures are unlikely to apply to it. When the held-out rows resemble the training rows, about one row in a hundred is flagged.

The chart shows one bar per held-out row in row order; the flagged rows are the bars at or above the line. With more than one target, the chart shows the model with the largest flagged share, and the figures name it.

The check treats inputs independently, so it will not notice a rare combination of values that are each ordinary on their own, such as a large height with a small weight. It is a first screen. A decision that depends on catching unusual inputs needs a dedicated check.

6.4 Fairness

This chapter checks the held-out predictions for uneven treatment of groups. For every categorical column with two to ten distinct values, other than the target, it reports per group the row count, mean actual value, mean prediction and mean residual, with R² and MAE for a number or accuracy, the share of rows classified correctly, for a yes-or-no target, and two measures across the groups.

  • Prediction parity ratio: the smallest group mean prediction divided by the largest, in absolute value. A ratio of 1.0 is perfect parity. Below 0.80, the four-fifths rule, some group receives systematically different predictions.
  • Largest mean residual: the largest absolute mean residual of any group. Above 30% of the overall residual standard deviation, at least one group is systematically over- or under-predicted.

The verdict combines the two: balanced when neither measure trips, parity gap or residual bias when one does, parity gap and residual bias when both do, and insufficient data when fewer than two groups have enough held-out rows.

Diafunc does not decide which columns are protected attributes; it reports every plausible one. A flagged column matters where it is a protected attribute in the decision the model supports, and can be set aside where it is not.

6.4.1 Rprice Fairness

Column Nearinc splits the held-out rows into 2 groups. The prediction-parity ratio is 0.95 (smallest over largest mean prediction), the largest absolute group residual mean is 21,376 (0.749 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. ldist goes closely with nearinc (association 0.84 on a scale from 0 to 1, and 4 more columns at 0.5 or above): a model that reads ldist can tell the groups of nearinc apart, so leaving nearinc out of the inputs does not keep it out of the predictions.

Aggregate residual std
28555
N columns
3
N proxies
10
Proxy threshold
0.5
Table 6.4.1 · Columns
Bias max abs residual meanColumnGroupsInterpretationN groupsN proxiesPrediction parity ratioVerdict
21375.7Nearincfalse, trueColumn Nearinc splits the held-out rows into 2 groups. The prediction-parity ratio is 0.95 (smallest over largest mean prediction), the largest absolute group residual mean is 21,376 (0.749 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. ldist goes closely with nearinc (association 0.84 on a scale from 0 to 1, and 4 more columns at 0.5 or above): a model that reads ldist can tell the groups of nearinc apart, so leaving nearinc out of the inputs does not keep it out of the predictions.250.95Residual bias
11217.7Y81true, falseColumn Y81 splits the held-out rows into 2 groups. The prediction-parity ratio is 0.976 (smallest over largest mean prediction), the largest absolute group residual mean is 11,218 (0.393 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. year goes closely with y81 (association 1.00 on a scale from 0 to 1, and 3 more columns at 0.5 or above): a model that reads year can tell the groups of y81 apart, so leaving y81 out of the inputs does not keep it out of the predictions.240.9758Residual bias
22802.7Y81nrincfalse, trueColumn Y81nrinc splits the held-out rows into 2 groups. The prediction-parity ratio is 0.992 (smallest over largest mean prediction), the largest absolute group residual mean is 22,803 (0.799 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. nearinc goes closely with y81nrinc (association 0.576 on a scale from 0 to 1): a model that reads nearinc can tell the groups of y81nrinc apart, so leaving y81nrinc out of the inputs does not keep it out of the predictions.210.9922Residual bias

6.5 Causal Honesty

Diafunc found a formula that captures the structure in the training data. That structure is correlative: it states what predicts what in this data. It does not establish that any input causes the target.

A correlative model is the right tool for predicting the target on new rows like these, for exploring which inputs carry information about the target, and for an early-stage diagnostic or screening tool.

A correlative model is not the right tool for claiming that intervening on an input will change the target, for attributing a target's values to specific causes for accountability, or for choosing between policies that act on the inputs.

The Quality by Input Range, Fairness and Concerns chapters show the most common confounding patterns in the data.

The model is correlative: it captures which columns predict the target in the training data, not whether changing one would change the other. Predictions are reliable for inputs that look like the training distribution; they are not evidence that intervening on an input will change the target. The 4 upgrade pathways below show what evidence would support a causal claim.

Table 6.5 · Upgrade pathways
NameSummary
Controlled experimentRandomize the input to be acted on, then analyze again. Randomization breaks confounding, so any change in the target is attributable to the input.
Natural experimentFind a setting where the input varied for reasons unrelated to the target, an instrumental variable. Variation of that kind supports causal inference.
Causal graphWrite out what plausibly causes what, then check whether the model's terms are consistent with the conditional independences the graph implies.
Control for confoundersIf a third variable plausibly moves both the input and the target, stratify by it or condition on it before training again. Gaps in the Quality by Input Range chapter are a clue.

6.6 Deployment Readiness

The analysis ran with the configuration below. The checks and the verdict of this chapter are read against it.

  • Targets: 1 target: rprice
  • Rows: 321 rows
  • Per-Segment Predictions: not produced, because no segment column was chosen
  • Fairness: compared across 3 categorical columns (nearinc, y81, y81nrinc); which of them are protected attributes is not decided here
  • Out-of-Distribution: held-out rows are scored against the training rows; predictions made outside this report are not scored
  • Provenance & Audit: recorded in the chapter of that name

Across 6 checks (3 pass, 3 warn, 0 fail), the model passes the hard gates but at least one check warns: Usable with care. A limited audience and monitoring of the warning signals come before a broad rollout.

Table 6.6 · Deployment Readiness
NameStatusDetail
GeneralizationPassquality on the held-out rows matches quality on the rows the model was found on
Prediction Interval CoverageWarntoo few held-out rows to judge coverage, largest gap 5.00 percentage points
Baseline Lift for RpriceWarnis not told apart from the best baseline, R² lift -0.227
Model ConstraintsPassevery held-out prediction is finite and within the training range
FairnessWarn3 columns compared, 3 flagged: Nearinc (a group is systematically over- or under-predicted), Y81 (a group is systematically over- or under-predicted), Y81nrinc (a group is systematically over- or under-predicted)
Out-of-DistributionPass1.6% of held-out rows are out of distribution
N critical concerns
0
N fail
0
N pass
3
N warn
3
Overall verdict
Caution

6.7 Provenance & Audit

This record captures the inputs and configuration that produced the report, so the analysis can be reproduced or the result defended in an audit.

Table 6.7 · Provenance & Audit
ItemValue
Analysis IDfe39ec60-48b7-11f1-8087-d378640fe787
Run time2026-11-07 14:25 UTC
Total rows321
Modeled targetsRprice
Segment columnnot available
Time limit1 h 0 min
Time taken1 h 7 min
Credit budget500 credits
Ufinq version0.3.58
Left out of the inputsLprice, Lrprice, Price

7 Data

Several variables move together tightly or form nearly perfect associations, especially between Year, Lintst, and Lintstsq. With 22 numeric columns and no columns with high missing values, the dataset supports precise checks for effects across all 321 rows.

Most notable in the Data section:

  • Y81 and Year go together (association 1.00)
  • Lintst and Lintstsq move together (correlation 0.999)
  • Y81ldist and Year move together (correlation 0.998)

How to read this

The Data section describes the table as supplied, before any modeling: the Column Profiles overview of every column and its role, one chapter per column with its distribution and correlations, a closer look at columns with notable characteristics, Duplicate Rows, Leakage Suspicion, a map for paired coordinate columns, and the General chapter with descriptive statistics, correlations, missing values, outlier ranges, clusters and segments across all columns.

7.1 Column Profiles

Profiled 25 columns across 321 rows (22 numeric, 3 non-numeric): every column is at least 75% populated. Distributional statistics on numeric columns can be read at face value. 6 columns are strongly skewed: the mean of such a column is pulled toward its long tail, so the median is the better guide to a typical value.

Table 7.1 · Columns
NameRoleTypeNMissingMeanStdMinP25MedianP75MaxSkewnessCV
RpriceTargetnumber321083,72133,11926,00059,00082,000100,230300,0001.3820.3956
AgeInputinteger321018.0132.57004221893.061.808
AgesqInputinteger321013824802001648435,7215.413.476
AreaInputinteger3210210769573515602056254451360.66760.3299
BathsInputinteger32102.3400.77112234-0.50250.3293
CbdInputinteger321015,82289671000900014,00023,00035,0000.087560.5667
DistInputinteger321020,7168508500013,40019,90027,20040,0000.082840.4107
IntstInputinteger321016,44290331000900016,00024,00034,0000.048320.5494
LandInputinteger321039,63039,514171016,93543,56046,100544,5007.680.9971
LareaInputnumber32107.5970.3416.6007.3527.6297.8418.544-0.3320.04485

The first 10 rows of 25. The other 15 are in the notebook.

N columns
25
N high missing
0
N rows
321
N strongly skewed
6
Table branch
main
Table ID
5744d005-bd93-41ce-a03c-84552051b259

How to read this

One row per column of the analyzed table: its role in the analysis (the target being predicted, an input it is predicted from, or the segment column), its type, value count and missing count, and for numeric columns the mean, standard deviation, minimum, quartiles, maximum and interquartile range. A column with 25% or more missing values is flagged; its mean and quartiles describe only the rows that have a value.

Two further figures describe the shape of a numeric column. The skewness says whether the values lean to one side: close to zero the column is symmetric, from 0.5 it is moderately skewed and from 1 strongly skewed, with the longer tail on the high side when the figure is positive and on the low side when it is negative. In a strongly skewed column the mean is pulled toward the tail, and the median is the better guide to a typical value. The coefficient of variation is the standard deviation divided by the mean: it compares how widely columns spread regardless of their units, and is given only for a column whose mean is positive.

These are the columns as supplied, before any modeling: the reference against which every later chapter reads.

7.2 Age

Age: 0 to 189, median 4, skewed to high values; moves with Agesq most (r 0.916).

How to read this

A column chapter profiles one column: its distribution in figures and charts, and how it relates to the other columns. It opens with one line on what the column holds. The Description chapter carries its figures, with every other column summarized for each of its values where it has few, and its association with every other column where it is numeric. The Visualization chapter carries its charts. A table with a date column adds an Over Time and a Time Series chapter; a target with classes adds a Class Balance chapter.

7.2.1 Description
Table 7.2.1 · Age
StatisticValue
Typeinteger
Values present321
Distinct values61
Continuousyes
Decimal places0
Minimum0
Maximum189
Mean18.01
Median4
Mode0

The first 10 rows of 17. The other 7 are in the notebook.

How to read this

The figures below summarize Age in numbers: where its values center (mean, median, mode), how far they spread (range, variance, standard deviation, interquartile range) and how they are distributed across percentiles. They are the reference for every chart in the Visualization chapter and for the correlation measures further down.

The figures for Age are grouped: the column itself first, then where its values center, how far they spread, the shape of their distribution and its percentiles. Each statistic is explained in the Description chapter under General in the Data section.

7.2.1.1 Correlation
Table 7.2.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Agesqtwo numbers10.91590.8963 to 0.93211< 0.001321
Bathstwo numbers0.6928-0.3569-0.4489 to -0.2575-0.6928-0.5842< 0.001321
Lrpricetwo numbers0.6762-0.416-0.5025 to -0.3211-0.6762-0.5182< 0.001321
Rpricetwo numbers0.6762-0.3319-0.4259 to -0.2308-0.6762-0.5182< 0.001321
Lpricetwo numbers0.6322-0.4013-0.4893 to -0.3053-0.6322-0.4731< 0.001321
Pricetwo numbers0.6322-0.3319-0.4259 to -0.2309-0.6322-0.4731< 0.001321
Cbdtwo numbers0.4992-0.3903-0.4793 to -0.2933-0.4992-0.3559< 0.001321
Intsttwo numbers0.4912-0.3892-0.4783 to -0.2922-0.4912-0.3508< 0.001321
Lintsttwo numbers0.4912-0.4256-0.5113 to -0.3316-0.4912-0.3508< 0.001321
Lintstsqtwo numbers0.4912-0.4248-0.5105 to -0.3307-0.4912-0.3508< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

How to read this

Each row pairs Age with one other column, numeric or categorical, and reports how strongly the two go together on an association from 0 to 1, and whether an association this strong could be chance. For another numeric column the row adds the direction: Pearson's coefficient measures a straight-line relationship, and Spearman's rho and Kendall's tau measure a monotonic relationship on ranks, which makes them resistant to outliers and lets them see a curved but consistently rising or falling pattern. For a categorical column the association is the correlation ratio η, which has no direction. A strong association is not an explanation: two columns can go together when a third one moves both.

The table lists every other numeric or categorical column against Age, with the pair's kind and association, three correlation coefficients and a confidence interval where the other column is numeric, the q-value and the rows the pair was measured on.

Table 7.2.1.1 · Correlation
MeasureMeaning
KindWhat the other column holds: a number (numeric) or categories (numeric_categorical). A categorical column is one that is not numeric and holds between 2 and 50 distinct values
AssociationThe strength of the pair from 0 to 1. With a numeric column: the stronger of Pearson's coefficient and Spearman's ρ, without its sign. With a categorical column: the correlation ratio η, the square root of the share of Age's spread that lies between the categories
CovarianceNumeric partner only. The direction of the relationship (positive or negative) in the units of the two columns. It is not standardized, so its size says little about strength without the columns' own spread
Pearson's Correlation CoefficientNumeric partner only. The strength and direction of a linear relationship, from -1 (a perfect falling line) through 0 (no linear relationship) to 1 (a perfect rising line)
Spearman's ρNumeric partner only. The strength and direction of a monotonic relationship, computed on ranks rather than values, from -1 to 1. It follows a curved pattern as long as it rises or falls consistently, and outliers move it little
Kendall's τNumeric partner only. The strength and direction of association from the agreement and disagreement of rank order across pairs of rows, from -1 to 1. It is the steadier of the two rank measures when the rows are few
q-valueHow likely a pair this strong is among unrelated columns, adjusted for the number of pairs involving Age (Benjamini-Hochberg). Among all pairs reported with q below 0.05, about 5% or fewer are expected to be false discoveries
95% CI (Pearson)Numeric partner only. The lower and upper bounds of the 95% confidence interval on the Pearson coefficient, from the Fisher z-transform. An interval that contains 0 leaves the direction of the relationship open
RowsThe rows in which both columns hold a value. The q-value and the interval of a pair rest on this count
7.2.2 Visualization

The rows of Age are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.05 between each value and the next), so the order carries no pattern to show.

How to read this

The charts in this chapter were chosen from the type of Age and of the columns it is compared with: a numeric column gets a histogram, a box plot and, when its row order carries a pattern, a value sequence, a categorical column gets a frequency chart, and its relation to each target of the analysis, and to the segment column when there is one, gets the chart that suits the pair, with the target on the vertical axis where it is a number. A column is not charted against every other column; which columns move together is shown under Strongest Correlations. The charts complement the figures in the Description chapter; the shape of a distribution and the position of its outliers are easier to see than to read.

7.3 Agesq

Agesq: 0 to 35,721, median 16, skewed to high values; moves with Age most (r 0.916).

7.3.1 Description
Table 7.3.1 · Agesq
StatisticValue
Typeinteger
Values present321
Distinct values61
Continuousyes
Decimal places0
Minimum0
Maximum35,721
Mean1381.6
Median16
Mode0

The first 10 rows of 17. The other 7 are in the notebook.

7.3.1.1 Correlation
Table 7.3.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Agetwo numbers10.91590.8963 to 0.93211< 0.001321
Bathstwo numbers0.6928-0.09382-0.2012 to 0.01581-0.6928-0.58420.112321
Lrpricetwo numbers0.6762-0.1711-0.2754 to -0.06278-0.6762-0.51820.007197321
Rpricetwo numbers0.6762-0.1103-0.2172 to -0.000851-0.6762-0.51820.06521321
Lpricetwo numbers0.6322-0.1858-0.2894 to -0.07792-0.6322-0.47310.00329321
Pricetwo numbers0.6322-0.1384-0.2442 to -0.02936-0.6322-0.47310.02854321
Cbdtwo numbers0.4992-0.191-0.2944 to -0.08331-0.4992-0.35590.003133321
Intsttwo numbers0.4912-0.1901-0.2934 to -0.0823-0.4912-0.35080.003133321
Lintsttwo numbers0.4912-0.1893-0.2927 to -0.08149-0.4912-0.35080.003133321
Lintstsqtwo numbers0.4912-0.1908-0.2942 to -0.0831-0.4912-0.35080.003133321

The first 10 rows of 24. The other 14 are in the notebook.

7.3.2 Visualization

The rows of Agesq are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.05 between each value and the next), so the order carries no pattern to show.

7.4 Area

Area: 735 to 5,136, median 2,056; moves with Larea most (r 0.974).

7.4.1 Description
Table 7.4.1 · Area
StatisticValue
Typeinteger
Values present321
Distinct values253
Continuousyes
Decimal places0
Minimum735
Maximum5136
Mean2106.7
Median2056
Mode2464

The first 10 rows of 17. The other 7 are in the notebook.

7.4.1.1 Correlation
Table 7.4.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Lareatwo numbers10.9740.9677 to 0.979111< 0.001321
Lrpricetwo numbers0.73380.66910.6039 to 0.72540.73380.5466< 0.001321
Rpricetwo numbers0.73380.67230.6075 to 0.72820.73380.5466< 0.001321
Lpricetwo numbers0.69970.64430.5754 to 0.70410.69970.5178< 0.001321
Pricetwo numbers0.69970.64530.5766 to 0.7050.69970.5178< 0.001321
Bathstwo numbers0.69440.66280.5966 to 0.720.69440.5625< 0.001321
Roomstwo numbers0.58210.53410.451 to 0.60810.58210.4675< 0.001321
Agetwo numbers0.4332-0.04542-0.1541 to 0.06437-0.4332-0.3090.4173321
Agesqtwo numbers0.43320.13280.02372 to 0.2388-0.4332-0.3090.0207321
Llandtwo numbers0.37210.37210.2738 to 0.46270.25640.169< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.4.2 Visualization

The rows of Area are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.03 between each value and the next), so the order carries no pattern to show.

7.5 Baths

Baths: 1 to 4, median 2, skewed to high values, 4 distinct values; moves with Lrprice most (r 0.746).

7.5.1 Description
Table 7.5.1 · Baths
StatisticValue
Typeinteger
Values present321
Distinct values4
Continuousyes
Decimal places0
Minimum1
Maximum4
Mean2.34
Median2
Mode3

The first 10 rows of 17. The other 7 are in the notebook.

7.5.1.1 By Value

Baths accounts for 49% of the spread in Rprice: 4 averages 81,293 above 1 (95% interval 13,472 to 149,113). An association, not a cause.

Table 7.5.1.1 · Groups
TargetValueRowsAverage or share95% interval
Rprice44131,11063126.1 to 199,095
Rprice3156105,366100,684 to 110,047
Rprice210667670.764314.7 to 71026.6
Rprice15549817.647157.1 to 52478.2
Table 7.5.1.1 · Mean per value of Baths
BathsRowsRoomsYearDistLrpricePriceLandAgesqAreaLintstCbdLareaY81ldistIntstLdistLlandNbhLpriceLintstsqAgeWindRprice
1555.6361979.515765.510.857361.420803.92038.31352.38.8379890.97.1764.65210290.99.5349.5592.32710.9378.9738.164.94549817.6
21066.349197918,80011.0975,377382422253.71886.49.3113849.17.4983.30114660.49.71110.172.61311.1887.3726.686.75567670.7
31567.0511979.523913.511.53122,47546540.3213.824769.82319365.47.7974.93119903.810.0410.641.93611.6696.673.2827.821105,366
447.751979.514,82511.74149,33965,75914782.539169.47811,5008.2594.75913,2509.57810.9211.8889.8685.58131,110

How to read this

The rows are grouped by the value of Baths, and every other column is summarized within each group with the same descriptive statistics. A statistic that shifts from group to group marks a column that moves with Baths; one that stays level marks a column that does not. Groups with few rows produce unstable figures, so the row count of each group is the first thing to read. Where the groups differ in a target by more than chance would give, the chapter states by how much; it states that the two go together, not that one brings the other about.

7.5.1.2 Correlation
Table 7.5.1.2 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Lrpricetwo numbers0.78570.74610.6933 to 0.7910.78570.6461< 0.001321
Rpricetwo numbers0.78570.68830.6259 to 0.74180.78570.6461< 0.001321
Lpricetwo numbers0.71660.67460.6102 to 0.73020.71660.5767< 0.001321
Pricetwo numbers0.71660.62590.5545 to 0.68820.71660.5767< 0.001321
Lareatwo numbers0.70930.70930.6504 to 0.75980.69440.5625< 0.001321
Areatwo numbers0.69440.66280.5966 to 0.720.69440.5625< 0.001321
Agetwo numbers0.6928-0.3569-0.4489 to -0.2575-0.6928-0.5842< 0.001321
Agesqtwo numbers0.6928-0.09382-0.2012 to 0.01581-0.6928-0.58420.1067321
Roomstwo numbers0.65380.60380.5293 to 0.6690.65380.5968< 0.001321
Llandtwo numbers0.49490.49490.4075 to 0.57330.37850.297< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.5.2 Visualization

The rows of Baths are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.01 between each value and the next), so the order carries no pattern to show.

7.6 Cbd

Cbd: 1,000 to 35,000, median 14,000, skewed to high values, 34 distinct values; moves with Intst most (r 0.987).

7.6.1 Description
Table 7.6.1 · Cbd
StatisticValue
Typeinteger
Values present321
Distinct values34
Continuousyes
Decimal places0
Minimum1000
Maximum35,000
Mean15822.4
Median14,000
Mode4000

The first 10 rows of 17. The other 7 are in the notebook.

7.6.1.1 By Value

Cbd accounts for 36% of the spread in Rprice: 11000 averages 80,963 above 5000 (95% interval 39,358 to 122,569). An association, not a cause.

Table 7.6.1.1 · Groups
TargetValueRowsAverage or share95% interval
Rprice1100012129,04888124 to 169,971
Rprice1200022112,10097769.4 to 126,431
Rprice2300018108,15598018.9 to 118,291
Rprice1400022104,97990162.8 to 119,796
Rprice220001395909.776736.2 to 115,083
Rprice16000995487.672694.5 to 118,281
Rprice200001193023.279326.7 to 106,720
Rprice240002493000.684135.1 to 101,866
Rprice15000392583.2-6364.7 to 191,531
Rprice13000690197.441576.2 to 138,819

The first 10 rows of 24. The other 14 are in the notebook.

Table 7.6.1.1 · Mean per value of Cbd
CbdRowsRoomsYearDistLrpricePriceLandAgesqAreaLintstLareaY81ldistIntstBathsLdistLlandLpriceLintstsqAgeWindRpriceNbh
1000261979.5880010.7354,5009464.56322.51634.58.157.3924.55235009.0829.02910.8666.4579.5346381.74
200085.8751978.8832510.8254,5008831.64217.31846.58.3667.492.21245001.59.019.00910.8870.0962.75351122.23.5
3000205.851979.2928010.856,65010986.22455.61594.78.1077.3253.66238501.69.1149.14310.9166.0442.63.750590.33.8
4000255.81979.110,01210.8961,56411675.61435.41683.78.2477.3623.34346401.689.1769.26310.9968.5131.684.455112.13.04
500055.81979.212,46010.7654,76010299.8513.413898.5327.2143.78952001.69.4299.17810.8772.8420.24.248084.52.4
60001261979.512858.310.9972945.815272.32589.71598.28.6677.3334.7415916.71.6679.469.4411.1275.1630.834.83361927.22.667
700056.41979.812,30010.9775,30028138.631621836.88.7047.4875.5864001.89.39310.0211.1375.8935.65.462125.21.6
800026.51979.510,40011.1479,00054081.5292.520189.197.5964.33310,0002.59.14110.7511.2784.516.5871113.7
900056.41979.217,06010.8864195.812,7897473.815678.9817.3293.9180001.69.7449.41710.9980.67533.855661
10000166.6251980.114162.511.22100,15029463.44502.62341.69.3077.6366.52511,1252.259.5139.9611.486.6442.38781782.4

The first 10 rows of 34. The other 24 are in the notebook.

7.6.1.2 Correlation
Table 7.6.1.2 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Intsttwo numbers0.98730.98730.9842 to 0.98980.98280.9179< 0.001321
Lintsttwo numbers0.98280.89790.8744 to 0.91720.98280.9179< 0.001321
Lintstsqtwo numbers0.98280.91860.8996 to 0.93410.98280.9179< 0.001321
Disttwo numbers0.95940.95940.9496 to 0.96720.95580.8512< 0.001321
Ldisttwo numbers0.95580.90580.884 to 0.92370.95580.8512< 0.001321
Nearinca number and a category0.7561< 0.001321
Windtwo numbers0.74330.66230.596 to 0.71960.74330.6059< 0.001321
Landtwo numbers0.73520.40190.3058 to 0.48980.73520.5485< 0.001321
Llandtwo numbers0.73520.72730.6713 to 0.77510.73520.5485< 0.001321
Agetwo numbers0.4992-0.3903-0.4793 to -0.2933-0.4992-0.3559< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.6.2 Visualization

The rows of Cbd are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.01 between each value and the next), so the order carries no pattern to show.

7.7 Dist

Dist: 5,000 to 40,000, median 19,900; moves with Ldist most (r 0.97).

7.7.1 Description
Table 7.7.1 · Dist
StatisticValue
Typeinteger
Values present321
Distinct values176
Continuousyes
Decimal places0
Minimum5000
Maximum40,000
Mean20715.6
Median19,900
Mode33300

The first 10 rows of 17. The other 7 are in the notebook.

7.7.1.1 Correlation
Table 7.7.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Ldisttwo numbers10.96960.9623 to 0.975611< 0.001321
Cbdtwo numbers0.95940.95940.9496 to 0.96720.95580.8512< 0.001321
Intsttwo numbers0.91660.91660.8971 to 0.93250.90880.7556< 0.001321
Lintsttwo numbers0.90880.81220.7713 to 0.84640.90880.7556< 0.001321
Lintstsqtwo numbers0.90880.83370.7969 to 0.86430.90880.7556< 0.001321
Nearinca number and a category0.7825< 0.001321
Landtwo numbers0.63760.33180.2307 to 0.42580.63760.4739< 0.001321
Llandtwo numbers0.63760.6340.5637 to 0.69520.63760.4739< 0.001321
Windtwo numbers0.57320.48970.4017 to 0.56870.57320.4489< 0.001321
Y81nrinca number and a category0.4595< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.7.2 Visualization

The rows of Dist are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.02 between each value and the next), so the order carries no pattern to show.

7.8 Intst

Intst: 1,000 to 34,000, median 16,000, 34 distinct values; moves with Cbd most (r 0.987).

7.8.1 Description
Table 7.8.1 · Intst
StatisticValue
Typeinteger
Values present321
Distinct values34
Continuousyes
Decimal places0
Minimum1000
Maximum34,000
Mean16442.4
Median16,000
Mode26000

The first 10 rows of 17. The other 7 are in the notebook.

7.8.1.1 By Value

Intst accounts for 36% of the spread in Rprice: 12000 averages 74,930 above 1000 (95% interval 30,733 to 119,127). An association, not a cause.

Table 7.8.1.1 · Groups
TargetValueRowsAverage or share95% interval
Rprice1200012122,03679119.2 to 164,954
Rprice1100012119,124100,714 to 137,534
Rprice160007111,77690618 to 132,933
Rprice1300018107,84587038.3 to 128,652
Rprice170008105,75972619 to 138,898
Rprice2400020102,88890988.7 to 114,788
Rprice200008101,81690199.6 to 113,433
Rprice230001897709.783990.8 to 111,429
Rprice250001197621.585522.1 to 109,721
Rprice140001095871.573622.8 to 118,120

The first 10 rows of 24. The other 14 are in the notebook.

Table 7.8.1.1 · Mean per value of Intst
IntstRowsRoomsYearDistLrpricePriceLandAgesqAreaLintstCbdLareaY81ldistBathsLdistLlandNbhLpriceLintstsqAgeWindRprice
1000561978.611,38010.7349,3806947.63847.81621.26.90838007.3421.8751.29.3398.802410.7847.7256.23.247106.9
2000115.9091979.410818.210.7653727.38083.13055.11766.57.6013272.77.4474.2161.7279.2878.882410.8857.7751.27347591.1
300056.21979.2930010.9665,00019,5936520.41867.28.00628007.5193.6081.69.1329.599411.0764.177.23.458690.9
4000125.6671979.311,17510.8760,90010746.12114.11541.18.2943666.77.2863.9051.4179.3159.247310.9868.7938.423.66753730.8
5000125.8331979.811583.311.0276204.215046.3572.61502.88.5174416.77.2515.4391.759.3449.566311.1772.5418.584.83363648.9
6000145.8571979.19692.910.8558207.110626.51819.11666.68.740007.393.3061.5719.1259.119410.9575.6833.214.28652143.3
7000165.8751978.910318.810.7754136.212421.12403.31585.48.8545312.57.3072.881.5639.1729.2671.7510.8678.3936.814.7549143.8
800056.21978.611,12011.1175,80024037.6278.61773.68.98762007.4211.87729.21310.0511.1680.7714.64.669769.3
9000481979.515,92511.1388,10020,29616757.32076.39.10587507.6284.8072.59.6659.73211.2782.991.755.573632.1
10000156.41980.617,74011.23103,46013946.71033.51845.79.2110733.37.4668.4892.29.7819.39311.4684.8320.6580815.4

The first 10 rows of 34. The other 24 are in the notebook.

7.8.1.2 Correlation
Table 7.8.1.2 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Lintsttwo numbers10.92840.9115 to 0.942111< 0.001321
Lintstsqtwo numbers10.94640.9336 to 0.956711< 0.001321
Cbdtwo numbers0.98730.98730.9842 to 0.98980.98280.9179< 0.001321
Disttwo numbers0.91660.91660.8971 to 0.93250.90880.7556< 0.001321
Ldisttwo numbers0.90880.85010.8167 to 0.87790.90880.7556< 0.001321
Windtwo numbers0.80240.72820.6724 to 0.77590.80240.6649< 0.001321
Landtwo numbers0.75870.42320.3289 to 0.5090.75870.5701< 0.001321
Llandtwo numbers0.75870.74850.6961 to 0.7930.75870.5701< 0.001321
Nearinca number and a category0.7186< 0.001321
Agetwo numbers0.4912-0.3892-0.4783 to -0.2922-0.4912-0.3508< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.8.2 Visualization

The rows of Intst are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.01 between each value and the next), so the order carries no pattern to show.

7.9 Land

Land: 1,710 to 544,500, median 43,560; moves with Lland most (r 0.709).

7.9.1 Description
Table 7.9.1 · Land
StatisticValue
Typeinteger
Values present321
Distinct values215
Continuousyes
Decimal places0
Minimum1710
Maximum544,500
Mean39629.9
Median43,560
Mode43560

The first 10 rows of 17. The other 7 are in the notebook.

7.9.1.1 Correlation
Table 7.9.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Llandtwo numbers10.70930.6504 to 0.759811< 0.001321
Windtwo numbers0.76380.44510.3528 to 0.52880.76380.6172< 0.001321
Intsttwo numbers0.75870.42320.3289 to 0.5090.75870.5701< 0.001321
Lintsttwo numbers0.75870.42850.3348 to 0.51390.75870.5701< 0.001321
Lintstsqtwo numbers0.75870.43340.34 to 0.51830.75870.5701< 0.001321
Cbdtwo numbers0.73520.40190.3058 to 0.48980.73520.5485< 0.001321
Disttwo numbers0.63760.33180.2307 to 0.42580.63760.4739< 0.001321
Ldisttwo numbers0.63760.32990.2287 to 0.42410.63760.4739< 0.001321
Lrpricetwo numbers0.3930.26530.1605 to 0.36420.3930.2536< 0.001321
Rpricetwo numbers0.3930.28650.1828 to 0.38390.3930.2536< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.9.2 Visualization

The rows of Land are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.03 between each value and the next), so the order carries no pattern to show.

7.10 Larea

Larea: 6.60 to 8.54, median 7.63; moves with Area most (r 0.974).

7.10.1 Description
Table 7.10.1 · Larea
StatisticValue
Typenumber
Values present321
Distinct values253
Continuousyes
Decimal places16
Minimum6.6
Maximum8.544
Mean7.597
Median7.629
Mode7.80954122543335

The first 10 rows of 17. The other 7 are in the notebook.

7.10.1.1 Correlation
Table 7.10.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Areatwo numbers10.9740.9677 to 0.979111< 0.001321
Lrpricetwo numbers0.73380.68910.6269 to 0.74260.73380.5466< 0.001321
Rpricetwo numbers0.73380.67110.6062 to 0.72720.73380.5466< 0.001321
Bathstwo numbers0.70930.70930.6504 to 0.75980.69440.5625< 0.001321
Lpricetwo numbers0.69970.65580.5886 to 0.7140.69970.5178< 0.001321
Pricetwo numbers0.69970.63970.5702 to 0.70020.69970.5178< 0.001321
Roomstwo numbers0.58210.56060.4806 to 0.63130.58210.4675< 0.001321
Agetwo numbers0.4332-0.09883-0.2061 to 0.01075-0.4332-0.3090.08404321
Agesqtwo numbers0.43320.08429-0.02542 to 0.192-0.4332-0.3090.1376321
Llandtwo numbers0.38770.38770.2905 to 0.47690.25640.169< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.10.2 Visualization

The rows of Larea are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.04 between each value and the next), so the order carries no pattern to show.

7.11 Ldist

Ldist: 8.52 to 10.6, median 9.90; moves with Dist most (r 0.97).

7.11.1 Description
Table 7.11.1 · Ldist
StatisticValue
Typenumber
Values present321
Distinct values176
Continuousyes
Decimal places15
Minimum8.517
Maximum10.6
Mean9.837
Median9.898
Mode10.413310050964355

The first 10 rows of 17. The other 7 are in the notebook.

7.11.1.1 Correlation
Table 7.11.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Disttwo numbers10.96960.9623 to 0.975611< 0.001321
Cbdtwo numbers0.95580.90580.884 to 0.92370.95580.8512< 0.001321
Intsttwo numbers0.90880.85010.8167 to 0.87790.90880.7556< 0.001321
Lintsttwo numbers0.90880.78080.7341 to 0.82020.90880.7556< 0.001321
Lintstsqtwo numbers0.90880.79860.7552 to 0.83510.90880.7556< 0.001321
Nearinca number and a category0.8404< 0.001321
Landtwo numbers0.63760.32990.2287 to 0.42410.63760.4739< 0.001321
Llandtwo numbers0.63760.63140.5607 to 0.6930.63760.4739< 0.001321
Windtwo numbers0.57320.47870.3897 to 0.55890.57320.4489< 0.001321
Y81nrinca number and a category0.4921< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.11.2 Visualization

The rows of Ldist are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.03 between each value and the next), so the order carries no pattern to show.

7.12 Lintst

Lintst: 6.91 to 10.4, median 9.68, skewed to low values, 34 distinct values; moves with Lintstsq most (r 0.999).

7.12.1 Description
Table 7.12.1 · Lintst
StatisticValue
Typenumber
Values present321
Distinct values34
Continuousyes
Decimal places15
Minimum6.908
Maximum10.43
Mean9.481
Median9.68
Mode10.166000366210938

The first 10 rows of 17. The other 7 are in the notebook.

7.12.1.1 By Value

Lintst accounts for 36% of the spread in Rprice: 9.3927001953125 averages 74,930 above 6.907800197601318 (95% interval 30,733 to 119,127). An association, not a cause.

Table 7.12.1.1 · Groups
TargetValueRowsAverage or share95% interval
Rprice9.392700195312512122,03679119.2 to 164,954
Rprice9.30570030212402312119,124100,714 to 137,534
Rprice9.6802997589111337111,77690618 to 132,933
Rprice9.47270011901855518107,84587038.3 to 128,652
Rprice9.7410001754760748105,75972619 to 138,898
Rprice10.08600044250488320102,88890988.7 to 114,788
Rprice9.9034996032714848101,81690199.6 to 113,433
Rprice10.0430002212524411897709.783990.8 to 111,429
Rprice10.1269998550415041197621.585522.1 to 109,721
Rprice9.5467996597290041095871.573622.8 to 118,120

The first 10 rows of 24. The other 14 are in the notebook.

Table 7.12.1.1 · Mean per value of Lintst
LintstRowsRoomsYearDistLrpricePriceLandAgesqAreaCbdLareaY81ldistIntstBathsLdistLlandNbhLpriceLintstsqAgeWindRprice
6.907800197601318561978.611,38010.7349,3806947.63847.81621.238007.3421.87510001.29.3398.802410.7847.7256.23.247106.9
7.600900173187256115.9091979.410818.210.7653727.38083.13055.11766.53272.77.4474.21620001.7279.2878.882410.8857.7751.27347591.1
8.00640010833740256.21979.2930010.9665,00019,5936520.41867.228007.5193.60830001.69.1329.599411.0764.177.23.458690.9
8.293999671936035125.6671979.311,17510.8760,90010746.12114.11541.13666.77.2863.90540001.4179.3159.247310.9868.7938.423.66753730.8
8.517200469970703125.8331979.811583.311.0276204.215046.3572.61502.84416.77.2515.43950001.759.3449.566311.1772.5418.584.83363648.9
8.69950008392334145.8571979.19692.910.8558207.110626.51819.11666.640007.393.30660001.5719.1259.119410.9575.6833.214.28652143.3
8.853699684143066165.8751978.910318.810.7754136.212421.12403.31585.45312.57.3072.8870001.5639.1729.2671.7510.8678.3936.814.7549143.8
8.98719978332519556.21978.611,12011.1175,80024037.6278.61773.662007.4211.877800029.21310.0511.1680.7714.64.669769.3
9.104999542236328481979.515,92511.1388,10020,29616757.32076.387507.6284.80790002.59.6659.73211.2782.991.755.573632.1
9.21030044555664156.41980.617,74011.23103,46013946.71033.51845.710733.37.4668.48910,0002.29.7819.39311.4684.8320.6580815.4

The first 10 rows of 34. The other 24 are in the notebook.

7.12.1.2 Correlation
Table 7.12.1.2 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Intsttwo numbers10.92840.9115 to 0.942111< 0.001321
Lintstsqtwo numbers10.99850.9982 to 0.998811< 0.001321
Cbdtwo numbers0.98280.89790.8744 to 0.91720.98280.9179< 0.001321
Disttwo numbers0.90880.81220.7713 to 0.84640.90880.7556< 0.001321
Ldisttwo numbers0.90880.78080.7341 to 0.82020.90880.7556< 0.001321
Windtwo numbers0.80240.72980.6742 to 0.77720.80240.6649< 0.001321
Llandtwo numbers0.79170.79170.7469 to 0.82930.75870.5701< 0.001321
Nearinca number and a category0.7687< 0.001321
Landtwo numbers0.75870.42850.3348 to 0.51390.75870.5701< 0.001321
Agetwo numbers0.4912-0.4256-0.5113 to -0.3316-0.4912-0.3508< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.12.2 Visualization

The rows of Lintst are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.04 between each value and the next), so the order carries no pattern to show.

7.13 Lintstsq

Lintstsq: 47.7 to 109, median 93.7, skewed to low values, 34 distinct values; moves with Lintst most (r 0.999).

7.13.1 Description
Table 7.13.1 · Lintstsq
StatisticValue
Typenumber
Values present321
Distinct values34
Continuousyes
Decimal places14
Minimum47.72
Maximum108.9
Mean90.48
Median93.71
Mode103.34756469726562

The first 10 rows of 17. The other 7 are in the notebook.

7.13.1.1 By Value

Lintstsq accounts for 36% of the spread in Rprice: 88.22281646728516 averages 74,930 above 47.71770477294922 (95% interval 30,733 to 119,127). An association, not a cause.

Table 7.13.1.1 · Groups
TargetValueRowsAverage or share95% interval
Rprice88.2228164672851612122,03679119.2 to 164,954
Rprice86.5960617065429712119,124100,714 to 137,534
Rprice93.708206176757817111,77690618 to 132,933
Rprice89.7320480346679718107,84587038.3 to 128,652
Rprice94.88708496093758105,75972619 to 138,898
Rprice101.7274017333984420102,88890988.7 to 114,788
Rprice98.079307556152348101,81690199.6 to 113,433
Rprice100.861854553222661897709.783990.8 to 111,429
Rprice102.55612945556641197621.585522.1 to 109,721
Rprice91.14138031005861095871.573622.8 to 118,120

The first 10 rows of 24. The other 14 are in the notebook.

Table 7.13.1.1 · Mean per value of Lintstsq
LintstsqRowsRoomsYearDistLrpricePriceLandAgesqAreaLintstCbdLareaY81ldistIntstBathsLdistLlandNbhLpriceAgeWindRprice
47.71770477294922561978.611,38010.7349,3806947.63847.81621.26.90838007.3421.87510001.29.3398.802410.7856.23.247106.9
57.773681640625115.9091979.410818.210.7653727.38083.13055.11766.57.6013272.77.4474.21620001.7279.2878.882410.8851.27347591.1
64.102439880371156.21979.2930010.9665,00019,5936520.41867.28.00628007.5193.60830001.69.1329.599411.0777.23.458690.9
68.7904281616211125.6671979.311,17510.8760,90010746.12114.11541.18.2943666.77.2863.90540001.4179.3159.247310.9838.423.66753730.8
72.5427017211914125.8331979.811583.311.0276204.215046.3572.61502.88.5174416.77.2515.43950001.759.3449.566311.1718.584.83363648.9
75.68130493164062145.8571979.19692.910.8558207.110626.51819.11666.68.740007.393.30660001.5719.1259.119410.9533.214.28652143.3
78.38800048828125165.8751978.910318.810.7754136.212421.12403.31585.48.8545312.57.3072.8870001.5639.1729.2671.7510.8636.814.7549143.8
80.7697601318359456.21978.611,12011.1175,80024037.6278.61773.68.98762007.4211.877800029.21310.0511.1614.64.669769.3
82.90101623535156481979.515,92511.1388,10020,29616757.32076.39.10587507.6284.80790002.59.6659.73211.2791.755.573632.1
84.82963562011719156.41980.617,74011.23103,46013946.71033.51845.79.2110733.37.4668.48910,0002.29.7819.39311.4620.6580815.4

The first 10 rows of 34. The other 24 are in the notebook.

7.13.1.2 Correlation
Table 7.13.1.2 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Intsttwo numbers10.94640.9336 to 0.956711< 0.001321
Lintsttwo numbers10.99850.9982 to 0.998811< 0.001321
Cbdtwo numbers0.98280.91860.8996 to 0.93410.98280.9179< 0.001321
Disttwo numbers0.90880.83370.7969 to 0.86430.90880.7556< 0.001321
Ldisttwo numbers0.90880.79860.7552 to 0.83510.90880.7556< 0.001321
Windtwo numbers0.80240.7380.6838 to 0.78410.80240.6649< 0.001321
Llandtwo numbers0.79550.79550.7514 to 0.83240.75870.5701< 0.001321
Nearinca number and a category0.7721< 0.001321
Landtwo numbers0.75870.43340.34 to 0.51830.75870.5701< 0.001321
Agetwo numbers0.4912-0.4248-0.5105 to -0.3307-0.4912-0.3508< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.13.2 Visualization

The rows of Lintstsq are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.04 between each value and the next), so the order carries no pattern to show.

7.14 Lland

Lland: 7.44 to 13.2, median 10.7, skewed to low values; moves with Lintstsq most (r 0.795).

7.14.1 Description
Table 7.14.1 · Lland
StatisticValue
Typenumber
Values present321
Distinct values215
Continuousyes
Decimal places15
Minimum7.444
Maximum13.21
Mean10.3
Median10.68
Mode10.681894302368164

The first 10 rows of 17. The other 7 are in the notebook.

7.14.1.1 Correlation
Table 7.14.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Landtwo numbers10.70930.6504 to 0.759811< 0.001321
Lintstsqtwo numbers0.79550.79550.7514 to 0.83240.75870.5701< 0.001321
Lintsttwo numbers0.79170.79170.7469 to 0.82930.75870.5701< 0.001321
Windtwo numbers0.76380.69360.6322 to 0.74640.76380.6172< 0.001321
Intsttwo numbers0.75870.74850.6961 to 0.7930.75870.5701< 0.001321
Cbdtwo numbers0.73520.72730.6713 to 0.77510.73520.5485< 0.001321
Disttwo numbers0.63760.6340.5637 to 0.69520.63760.4739< 0.001321
Ldisttwo numbers0.63760.63140.5607 to 0.6930.63760.4739< 0.001321
Nearinca number and a category0.6122< 0.001321
Lrpricetwo numbers0.54690.54690.4653 to 0.61930.3930.2536< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.14.2 Visualization

The rows of Lland are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.02 between each value and the next), so the order carries no pattern to show.

7.15 Lprice

The model does not use this column: it is left out of the model's inputs. Its figures and charts describe the data all the same, and its relation to the target shows what the model is not told.

Lprice: 10.2 to 12.6, median 11.4; moves with Price most (r 0.962).

7.15.1 Description
Table 7.15.1 · Lprice
StatisticValue
Typenumber
Values present321
Distinct values200
Continuousyes
Decimal places15
Minimum10.17
Maximum12.61
Mean11.38
Median11.36
Mode11.695246696472168

The first 10 rows of 17. The other 7 are in the notebook.

7.15.1.1 Correlation
Table 7.15.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Pricetwo numbers10.96150.9523 to 0.96911< 0.001321
Lrpricetwo numbers0.95670.95670.9464 to 0.96510.95540.8548< 0.001321
Rpricetwo numbers0.95540.92130.9029 to 0.93630.95540.8548< 0.001321
Bathstwo numbers0.71660.67460.6102 to 0.73020.71660.5767< 0.001321
Areatwo numbers0.69970.64430.5754 to 0.70410.69970.5178< 0.001321
Lareatwo numbers0.69970.65580.5886 to 0.7140.69970.5178< 0.001321
Agetwo numbers0.6322-0.4013-0.4893 to -0.3053-0.6322-0.4731< 0.001321
Agesqtwo numbers0.6322-0.1858-0.2894 to -0.07792-0.6322-0.4731< 0.001321
Y81ldisttwo numbers0.56950.52530.4412 to 0.60020.56950.4144< 0.001321
Roomstwo numbers0.55910.49330.4058 to 0.57190.55910.4391< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.15.2 Visualization

The rows of Lprice are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.04 between each value and the next), so the order carries no pattern to show.

7.16 Lrprice

The model does not use this column: it is left out of the model's inputs. Its figures and charts describe the data all the same, and its relation to the target shows what the model is not told.

Lrprice: 10.2 to 12.6, median 11.3; moves with Rprice most (r 0.959).

7.16.1 Description
Table 7.16.1 · Lrprice
StatisticValue
Typenumber
Values present321
Distinct values236
Continuousyes
Decimal places15
Minimum10.17
Maximum12.61
Mean11.26
Median11.31
Mode10.778956413269043

The first 10 rows of 17. The other 7 are in the notebook.

7.16.1.1 Correlation
Table 7.16.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Rpricetwo numbers10.95920.9494 to 0.967111< 0.001321
Lpricetwo numbers0.95670.95670.9464 to 0.96510.95540.8548< 0.001321
Pricetwo numbers0.95540.91470.8949 to 0.9310.95540.8548< 0.001321
Bathstwo numbers0.78570.74610.6933 to 0.7910.78570.6461< 0.001321
Areatwo numbers0.73380.66910.6039 to 0.72540.73380.5466< 0.001321
Lareatwo numbers0.73380.68910.6269 to 0.74260.73380.5466< 0.001321
Agetwo numbers0.6762-0.416-0.5025 to -0.3211-0.6762-0.5182< 0.001321
Agesqtwo numbers0.6762-0.1711-0.2754 to -0.06278-0.6762-0.51820.002099321
Roomstwo numbers0.61840.55530.4747 to 0.62670.61840.4878< 0.001321
Llandtwo numbers0.54690.54690.4653 to 0.61930.3930.2536< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.16.2 Visualization

The rows of Lrprice are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.03 between each value and the next), so the order carries no pattern to show.

7.17 Nbh

Nbh: 0 to 6, median 2, 7 distinct values.

7.17.1 Description
Table 7.17.1 · Nbh
StatisticValue
Typeinteger
Values present321
Distinct values7
Continuousyes
Decimal places0
Minimum0
Maximum6
Mean2.209
Median2
Mode0

The first 10 rows of 17. The other 7 are in the notebook.

7.17.1.1 By Value

Nbh accounts for 19% of the spread in Rprice: 1 averages 42,654 above 4 (95% interval 27,741 to 57,567). An association, not a cause.

Table 7.17.1.1 · Groups
TargetValueRowsAverage or share95% interval
Rprice12796376.981823.3 to 110,930
Rprice63293947.985263.1 to 102,633
Rprice012192454.485344.2 to 99564.6
Rprice24786454.180687.6 to 92220.7
Rprice52784423.578400.2 to 90446.8
Rprice377327049749.4 to 96790.6
Rprice46053723.250214.2 to 57232.3
Table 7.17.1.1 · Mean per value of Nbh
NbhRowsRoomsYearDistLrpricePriceLandAgesqAreaLintstCbdLareaY81ldistIntstBathsLdistLlandLpriceLintstsqAgeWindRprice
01216.621979.619647.911.35108,73733739.31389.321709.45614041.37.6295.20214247.92.4219.81610.2311.4989.6415.775.97592454.4
1276.8891979.315166.711.41109,80075870.31489.32408.29.75114888.97.7124.14217481.52.6679.56210.9911.5395.1113.631196376.9
2476.872197926591.511.3496102.850677.8440.52113.59.96921425.57.6133.47821638.32.53210.1810.811.4399.419.4047.25586454.1
376.4291979.334,30011.1585414.355984.984.571869.410.2128,0007.4824.47227285.72.42910.4410.8911.26104.37.714773270
4605.9331979.210,41010.8660637.511116.42483.51669.98.2213766.77.373.7084266.71.6179.2259.16110.9767.9141.174.01753723.2
5276.6671978.934229.611.3393125.956624.622.932095.610.3330222.27.6213.09830740.72.6310.4410.8511.41106.72.7049.07484423.5
6326.9691979.425753.111.42108,09750,6452007.42483.410.1122906.37.7884.74424,6252.56310.1510.811.54102.214.5610.7593947.9
7.17.1.2 Correlation
Table 7.17.1.2 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Windtwo numbers0.24080.23380.1275 to 0.33470.24080.1879< 0.001321
Intsttwo numbers0.22790.21150.1045 to 0.31370.22790.1750.001075321
Lintsttwo numbers0.22790.003938-0.1056 to 0.11340.22790.1750.9617321
Lintstsqtwo numbers0.22790.02956-0.08017 to 0.13860.22790.1750.6831321
Pricetwo numbers0.2159-0.2159-0.3179 to -0.109-0.1902-0.13630.001075321
Rpricetwo numbers0.2076-0.2076-0.31 to -0.1004-0.1757-0.12510.001078321
Cbdtwo numbers0.19590.18810.08023 to 0.29150.19590.14930.002834321
Lpricetwo numbers0.1902-0.1891-0.2925 to -0.08132-0.1902-0.13630.002834321
Lrpricetwo numbers0.1796-0.1796-0.2835 to -0.07152-0.1757-0.12510.004225321
Disttwo numbers0.17070.17070.06243 to 0.27510.16370.11250.006433321

The first 10 rows of 24. The other 14 are in the notebook.

7.17.2 Visualization

The rows of Nbh are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.04 between each value and the next), so the order carries no pattern to show.

7.18 Nearinc

Nearinc: 2 distinct values, most often no.

7.18.1 Description
Table 7.18.1 · Nearinc
StatisticValue
Typebool
Values present321
Distinct values2
Continuousno
Modefalse
Table 7.18.1 · Rows per value
VariableValueRowsShare
Nearincfalse22570.09%
Nearinctrue9629.91%
7.18.1.1 By Value

Nearinc accounts for 11% of the spread in Rprice: no averages 24,457 above yes (95% interval 15,717 to 33,197). An association, not a cause.

Table 7.18.1.1 · Groups
TargetValueRowsAverage or share95% interval
Rpricefalse22591035.587502.9 to 94568.1
Rpricetrue9666578.858560.4 to 74597.3
Table 7.18.1.1 · Mean per value of Nearinc
NearincRowsRoomsYearDistLrpricePriceLandAgesqAreaLintstCbdLareaY81ldistIntstBathsLdistLlandNbhLpriceLintstsqAgeWindRprice
false2256.81979.425057.811.37104,90546984.9969.52200.19.8720244.47.6554.55720675.62.55110.110.622.03111.4997.5710.827.74791035.5
true966.0831979.310538.51175465.122391.62347.31887.88.5675458.37.4613.8396520.81.8449.2239.5522.62511.1173.8834.855.17766578.8

7.19 Price

The model does not use this column: it is left out of the model's inputs. Its figures and charts describe the data all the same, and its relation to the target shows what the model is not told.

Price: 26,000 to 300,000, median 85,900, skewed to high values; moves with Lprice most (r 0.962).

7.19.1 Description
Table 7.19.1 · Price
StatisticValue
Typeinteger
Values present321
Distinct values200
Continuousyes
Decimal places0
Minimum26,000
Maximum300,000
Mean96100.7
Median85,900
Mode120000

The first 10 rows of 17. The other 7 are in the notebook.

7.19.1.1 Correlation
Table 7.19.1.1 · Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Lpricetwo numbers10.96150.9523 to 0.96911< 0.001321
Lrpricetwo numbers0.95540.91470.8949 to 0.9310.95540.8548< 0.001321
Rpricetwo numbers0.95540.95170.9402 to 0.9610.95540.8548< 0.001321
Bathstwo numbers0.71660.62590.5545 to 0.68820.71660.5767< 0.001321
Areatwo numbers0.69970.64530.5766 to 0.7050.69970.5178< 0.001321
Lareatwo numbers0.69970.63970.5702 to 0.70020.69970.5178< 0.001321
Agetwo numbers0.6322-0.3319-0.4259 to -0.2309-0.6322-0.4731< 0.001321
Agesqtwo numbers0.6322-0.1384-0.2442 to -0.02936-0.6322-0.47310.01365321
Y81ldisttwo numbers0.56950.52050.4359 to 0.59610.56950.4144< 0.001321
Roomstwo numbers0.55910.44310.3507 to 0.5270.55910.4391< 0.001321

The first 10 rows of 24. The other 14 are in the notebook.

7.19.2 Visualization

The rows of Price are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.04 between each value and the next), so the order carries no pattern to show.

7.20 Rooms

Rooms: 4 to 10, median 7, skewed to low values, 7 distinct values; moves with Baths most (r 0.604).

7.20.1 Description
Table 7.20.1 · Rooms
StatisticValue
Typeinteger
Values present321
Distinct values7
Continuousyes
Decimal places0
Minimum4
Maximum10
Mean6.586
Median7
Mode7

The first 10 rows of 17. The other 7 are in the notebook.

7.20.1.1 By Value

Rooms accounts for 33% of the spread in Rprice: 9 averages 70,861 above 4 (95% interval 47,311 to 94,412). An association, not a cause.

Table 7.20.1.1 · Groups
TargetValueRowsAverage or share95% interval
Rprice911114,35191302.1 to 137,400
Rprice716998892.494176.2 to 103,609
Rprice81085652.865732.1 to 105,574
Rprice69664973.660451.6 to 69495.6
Rprice53052300.848247.6 to 56354
Rprice4443489.834218.3 to 52761.3
Table 7.20.1.1 · Mean per value of Rooms
RoomsRowsYearDistLrpricePriceLandAgesqAreaLintstCbdLareaY81ldistIntstBathsLdistLlandNbhLpriceLintstsqAgeWindRprice
441978.814,07510.6747,37517,666747.312498.72670007.0342.34677509.5069.493310.7476.5326.75543489.8
5301979.213,17010.8458732.616183.51633.71414.18.73975007.2063.7438333.31.49.3729.3152.13310.9576.9434.334.66752300.8
6961979.318760.411.0374415.138331.81682.11748.69.24313593.87.4264.34914281.31.8659.71110.132.52111.1586.2626.546.4964973.6
71691979.423485.811.46114,21544605.94812384.19.75718881.77.7494.59119349.12.7571010.571.99411.5895.56.1837.59298892.4
8101979.219,73011.3198,99032,8601568.92331.79.4114,9007.6924.00315,9002.49.79310.233.611.4189.2622.7785652.8
9111979.119436.411.61126,71855060.192122999.49.69615545.57.9553.52917090.93.0919.81810.771.63611.794.1258.739114,351
101197817,10011.1166,90012,51631,68417509.10510,0007.4670900029.7479.43511.1182.9178566,900
7.20.1.2 Correlation

This section continues in the notebook.

The report continues in the notebook.

Loading Diafunc…