Analysis report

Car Seat Sales

Data from R package ISLR (CRAN), published under https://www.gnu.org/licenses/old-licenses/gpl-2.0.html.

Target
Sales
Analyzed
2026-10-23 22:00 UTC

Summary

The Sales model uses 5 of the 10 available inputs and relies most heavily on Price: as Price rises across its observed range, the model’s output falls. For this input, the analysis finds that moving Price from 99.0 to 130 lowers the model’s output by about 2.80 on average, with every other input as observed. The model’s answer stands out from simpler approaches by consistently adding a modest improvement over what a straight line through the data would capture, being closer to the true values on nearly two-thirds of the rows tested.

The model leaves out half the available columns and, while it tracks most rows well, it predicts Sales much less accurately when ShelveLoc is Bad. It also performs less strongly for high Advertising, high Income, and non-US rows. The model’s formula is short but uses specialist functions, so interpretation for decision-making may require technical support. Before operational use, validate its performance on a new sample, with special attention to areas where accuracy was lower.

Formula

Sales = 6.9 + 0.11 × Advertising − 0.039 × Age + 0.0934 × CompPrice − 0.0903 × Price + 3.38 × indicator(ShelveLoc = "Good")Sales=6.9+0.11×Advertising−0.039×Age+0.0934×CompPrice−0.0903×Price+3.38×ShelveLoc⩵Good
Sales = 6.9 + 0.11 × Advertising − 0.039 × Age + 0.0934 × CompPrice − 0.0903 × Price + 3.38 × indicator(ShelveLoc = "Good")
6 terms · additively decomposable

Held-out diagnostics

TargetQualityErrorR²MAEGeneralizationRows
Sales70.95 %25.31 %0.74691.078generalizing80
  • Sales · Quality 95 % CI 60.7 % – 78.58 %
  • Sales · Error 95 % CI 18.66 % – 34.24 %
  • Sales · R² 95 % CI 0.6576 – 0.8134
  • Sales · MAE 95 % CI 0.9133 – 1.266

Concerns

WARNING implications

Weaker quality on 4 slices of the data

On 4 of the 34 slices compared the model is weaker than on all held-out rows (R² 0.747). The largest gaps: ShelveLoc = Bad (R² 0.105 against 0.747 on all rows, on 24 rows) and Income in the highest quarter (94 to 120) (R² 0.575 against 0.747 on all rows, on 21 rows). Predictions for these rows deserve caution; the Quality by Input Range chapter lists every slice.

Complete list (4)
SegmentRowsR²Against all rowsVerdict
ShelveLoc = Bad240.105-0.642Much weaker
Income in the highest quarter (94 to 120)210.575-0.172Weaker
Advertising in the highest quarter (12 to 29)230.618-0.129Weaker
US = no350.644-0.103Weaker

INFO implications

If `US` is a characteristic predictions must not follow, `Advertising` stands in for it

This matters only if US is a characteristic the predictions must not follow, such as a person's age, sex or origin; for an ordinary property of the data it needs no action. Advertising and US go together with an association of 0.684 on a scale from 0 to 1. The model reads Advertising. A model that reads Advertising can tell the groups of US apart, so leaving US out of the inputs would not keep it out of the predictions. The Fairness chapter compares the model across its groups. An association, not a cause.

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 2026-10-23 22:00 UTC, analysis f1ec5b50-48b7-11f1-8087-d378640fe787, 1 target, time budget 1 h 0 min, compute budget 500 credits, 400 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 very high quality and still tell you nothing new.

What it found

Sales starts from a base of 6.9, rises with CompPrice and Advertising, falls with Price and Age, and also changes with ShelveLoc.

  1. Price carries most of the model. Price accounts for 63% of what the formula explains and ShelveLoc for 17%. CompPrice adds 12%, Advertising 5% and Age 3.8%.
  2. Sales falls with Price and rises with CompPrice. For each additional 10 units of Price, the model's output for Sales is lower by 0.903. For each additional unit of CompPrice, it is higher by 0.0934. Each holds with the other inputs at typical values.
  3. ShelveLoc accounts for 31% of the spread in Sales. Good averages 4.69 above Bad (95% interval 3.98 to 5.41).
  4. The formula is ahead of a straight line. The formula explains 75% of Sales; a straight line explains 54%. That was measured on 80 rows it never saw. The formula typically misses Sales by 1.08. It is weakest where ShelveLoc is Bad, where it explains 11%.

Summary

What to do next: Intervals miss more rows than their level allows, largest gap 5.00 percentage points.

5 chapters, each headed by what it found.

Price carries most of the formula

What the formula relies on. Shares of the formula: Price 63%, ShelveLoc 17%, CompPrice 12%.

The model's output falls with Price on every one of the 10,240 steps between observed values. The model's output rises with CompPrice on every one of the 9,440 steps between observed values. The model's output rises with Advertising on every one of the 3,680 steps between observed values. Moving Price from 99.0 to 130, its first and third quartile, with every other input as observed lowers the model's output by 2.80 on average. These are measurements on the rows the search trained on, inside the observed ranges; outside them nothing is claimed.

InputShare of importancePermutation importanceSensitivityResidual correlationOccurrencesModel's output as the input risesChange, first to third quartile
Price62.8 %5.7730.09030.26031Falls (10,240 of 10,240 steps)-2.799 (99 to 130)
ShelveLoc16.64 %1.5291
CompPrice11.8 %1.0840.09340.26961Rises (9440 of 9440 steps)+1.868 (113 to 133)
Advertising5.01 %0.4610.11-0.13841Rises (3680 of 3680 steps)+1.21 (0 to 11)
Age3.75 %0.3450.039-0.088851Falls (8320 of 8320 steps)-0.975 (41 to 66)
Rows the search trained on, measured
160
Inputs the formula does not read
Education, Income, Population, Urban, US

Read as a sum, the formula has 6 additive terms: 5 that read an input and a constant base. All that read an input are needed to reach 100% of the variation of the model's output on the 160 rows the search trained on: the largest are -0.0903 × Price (43%), 3.38 × indicator(ShelveLoc = "Good") (37%) and 0.11 × Advertising (7.2%). A share describes the model's output, not the target. Each curve below shows one term's response to its free inputs, with the other inputs held at the training median.

Additive
Yes
Rows the search trained on, measured
160
Terms in the formula
6
Notation
6.9 + 0.11 × Advertising − 0.039 × Age + 0.0934 × CompPrice − 0.0903 × Price + 3.38 × indicator(ShelveLoc = "Good")
Target
Sales

It holds on new rows: R² 0.75 held out, 0.78 on the training rows

How well it holds. The formula explains 75% of Sales; a straight line explains 54%.

R² of 0.747: good. The model captures most of the variation; individual predictions still carry uncertainty; the Prediction Interval Coverage chapter shows how wide it is. Mean absolute error is 1.08 on the target's scale.

Error
25.31 % (95 % CI 18.66 % – 34.24 %)
MAE
1.078 (95 % CI 0.9133 – 1.266)
N
80
Training rows
320
Quality
70.95 % (95 % CI 60.7 % – 78.58 %)
R²
0.7469 (95 % CI 0.6576 – 0.8134)
Target
Sales

The model beats the strongest simple yardstick, a straight line through the data, by 21 points of variance explained on the held-out rows. Row by row the margin is small: on many individual rows the yardstick does about as well, and whether the complexity is justified depends on the use. Of the 80 held-out rows where the two differ, the model is closer on 65% (95% interval 54% to 75%).

Best yardstick's score
0.5407 (95 % CI 0.3823 – 0.6599)
Best baseline R²
0.5407
Effect size d
0.2942 (95 % CI 0.09685 – 0.5146)
Effect size d magnitude
small
Margin over the best yardstick
0.2061 (95 % CI 0.1005 – 0.3384)
Margin's interval stays above zero
Yes
Model MAE
1.078 (95 % CI 0.9133 – 1.266)
Model
0.7469 (95 % CI 0.6576 – 0.8134)
Model R²
0.7469 (95 % CI 0.6576 – 0.8134)
Held-out rows
80
Held-out rows without an answer
0
Target
Sales
Rows where the model is closer
65 % (95 % CI 54.08 % – 74.55 %)
Rows compared
80
Rows level
0
Regression yardstick trained with gaps filled (too few complete rows)
No
Training rows of the regression yardstick
320

The model explains held-out rows about as well as training rows (held-out R² 0.747, gap between training and held-out R² 0.0287): it generalizes. The structure the formula encodes carries from the training rows to the held-out rows.

Generalization gap
0.02867
N test
80
Target
Sales
Test error
1.078 (95 % CI 0.9133 – 1.266)
Test R²
0.7469 (95 % CI 0.6576 – 0.8134)
Train error
0.1803
Train R²
0.7755
Verdict
Generalizing

Weaker on 4 slices of the data, most in ShelveLoc = Bad

Where it is weak. The largest miss on a held-out row is 3.97 (recorded 3.07, predicted 7.04). 5% of the held-out rows are unlike the rows it learned from.

Segment shelveloc = Bad is much weaker than the aggregate (24 held-out rows): segment R² 0.105 against aggregate R² 0.747 (gap -0.642). A drop this large means the model describes this slice of the data poorly; an input interaction or a missing column may be specific to this subset.

SegmentRowsR²Gap to all rowsVerdict
ShelveLoc = Bad240.1052-0.64167Much weaker than on all rows
Income = q4210.5752-0.17167Weaker than on all rows
Advertising = q4230.6176-0.12930Weaker than on all rows
US = false350.6436-0.10329Weaker than on all rows
Price = q1210.6508-0.09607In line with all rows
Age = q4220.6810-0.06583In line with all rows
Advertising = q2430.6826-0.06424In line with all rows
Population = q2210.7038-0.04307In line with all rows
Price = q3220.7068-0.04012In line with all rows
Education = q2230.7363-0.01056In line with all rows

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

Aggregate R²
0.7469
Measure
r2
Target
Sales

The worst held-out row was missed by 3.97 (target's scale). The 10 rows with the largest absolute residuals out of 80 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.

ActualPredictedResidualAdvertisingAgeCompPricePriceShelveLoc
3.0707.035-3.9651375118104Bad
4.1507.359-3.209338141128Bad
1.8204.870-3.050077139133Bad
3.1305.939-2.8091166130130Bad
12.1109.496+2.6141826118104Medium
7.50010.082-2.58204314091Bad
3.5805.926-2.346072142164Good
8.9706.654+2.316033132125Medium
10.5908.418+2.1721530131124Medium
10.3608.247+2.1131834107103Medium
Additive
Yes
N extrapolated
0
Held-out rows
80
Held-out rows for the interval
80
Residual P5
-2.593
Residual P95
1.943
Residual std
1.38

4 of 80 held-out rows (5%) lie further from the training rows than 99 in 100 rows like them would, measured on the 4 inputs the model reads (advertising, age, compprice, price). That is a small minority; the held-out figures describe the model on rows like the ones it was trained on.

TargetInputs measuredLineHeld-out rowsBeyond the lineShare
SalesAdvertising, Age, CompPrice, Price3.6448045 %
Score bins · 24 of 30 rows shown
CountFromTo
000.175
00.1750.35
10.350.525
10.5250.7
20.70.875
30.8751.05
21.051.225
41.2251.4
71.41.575
111.5751.75

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

Not ready to use yet: 2 checks failed

What to watch. First: Weaker quality on 4 slices of the data.

SeveritySectionConcernDetail
WarningImplicationsWeaker quality on 4 slices of the dataOn 4 of the 34 slices compared the model is weaker than on all held-out rows (R² 0.747). The largest gaps: ShelveLoc = Bad (R² 0.105 against 0.747 on all rows, on 24 rows) and Income in the highest quarter (94 to 120) (R² 0.575 against 0.747 on all rows, on 21 rows). Predictions for these rows deserve caution; the Quality by Input Range chapter lists every slice.
InfoImplicationsIf US is a characteristic predictions must not follow, Advertising stands in for itThis matters only if US is a characteristic the predictions must not follow, such as a person's age, sex or origin; for an ordinary property of the data it needs no action. Advertising and US go together with an association of 0.684 on a scale from 0 to 1. The model reads Advertising. A model that reads Advertising can tell the groups of US apart, so leaving US out of the inputs would not keep it out of the predictions. The Fairness chapter compares the model across its groups. An association, not a cause.

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

  • Targets: 1 target: sales
  • Rows: 400 rows
  • Per-Segment Predictions: not produced, because no segment column was chosen
  • Fairness: compared across 3 categorical columns (shelveloc, urban, us); 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 (4 pass, 0 warn, 2 fail), at least one check fails or a critical concern is open: Not ready. The failing checks need addressing first; they typically indicate data leakage, severe miscalibration, or a model that loses to its baseline.

NameStatusDetail
GeneralizationPassheld-out quality holds up against training quality
Prediction Interval CoverageFailintervals miss more rows than their level allows, largest gap 5.00 percentage points
Baseline Lift for SalesPassbeats the best baseline, R² lift 0.206
Model ConstraintsPassevery held-out prediction is finite and within the training range
FairnessFail3 columns compared, 1 flagged: ShelveLoc (a group both receives smaller predictions and is systematically mis-predicted)
Out-of-DistributionPass5% of held-out rows are out of distribution
N critical concerns
0
N fail
2
N pass
4
N warn
0
Overall verdict
Not ready

Column ShelveLoc splits the held-out rows into 3 groups. The prediction-parity ratio is 0.733 (smallest over largest mean prediction), the largest absolute group residual mean is 1.28 (0.929 times the overall residual standard deviation). Parity gap and residual bias: the model both predicts systematically lower for some groups and errs systematically against at least one group. This is the worst-case fairness signal on this column; review before deployment.

Aggregate residual std
1.38
N columns
3
N proxies
1
Proxy threshold
0.5
Columns
Bias max abs residual meanColumnGroupsInterpretationN groupsN proxiesPrediction parity ratioVerdict
1.282ShelveLocMedium, Good, BadColumn ShelveLoc splits the held-out rows into 3 groups. The prediction-parity ratio is 0.733 (smallest over largest mean prediction), the largest absolute group residual mean is 1.28 (0.929 times the overall residual standard deviation). Parity gap and residual bias: the model both predicts systematically lower for some groups and errs systematically against at least one group. This is the worst-case fairness signal on this column; review before deployment.30.7334Parity and bias
0.161Urbanfalse, trueColumn Urban splits the held-out rows into 2 groups. The prediction-parity ratio is 0.983 (smallest over largest mean prediction), the largest absolute group residual mean is 0.161 (0.117 times the overall residual standard deviation). Balanced: every group is within the 80% parity band and no group's residual mean exceeds 30% of the overall residual standard deviation. No systematic disparity was detected on this column.20.9828Balanced
0.1476UStrue, falseColumn US splits the held-out rows into 2 groups. The prediction-parity ratio is 0.926 (smallest over largest mean prediction), the largest absolute group residual mean is 0.148 (0.107 times the overall residual standard deviation). Balanced: every group is within the 80% parity band and no group's residual mean exceeds 30% of the overall residual standard deviation. No systematic disparity was detected on this column. advertising goes closely with us (association 0.684 on a scale from 0 to 1): a model that reads advertising can tell the groups of us apart, so leaving us out of the inputs does not keep it out of the predictions.210.9262Balanced

More rows in ShelveLoc = Bad would help most

What to collect next. Its R² there is 0.11 against 0.75 on all rows, on 24 held-out rows.

  • Collect more rows where ShelveLoc = Bad. Reason: the model is far less accurate on the 24 held-out rows in this segment, explaining 11% of the variation there.
  • Collect more rows where Income is in the highest quarter (94 to 120), Advertising is in the highest quarter (12 to 29), and US = no. Reason: the model is less accurate than overall on these segments, explaining 58% to 64% of the variation on 21 to 35 held-out rows in each.
  • Inspect predictions for ShelveLoc for parity and bias. Reason: a fairness flag was raised for the column ShelveLoc with the verdict parity and bias.

Evidence

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

Headline

Diafunc modeled Sales from 10 columns, with the model using 5. The model explains 75% of the variation in Sales and relies most on Price, with which the model's output falls.

  • One target was modeled; 80 rows were held out for checking.
  • A straight line through the data already explained 54%; the model added 21 points of variance explained, a small but consistent margin, and was closer on 65% of the rows checked.
  • The model is far less accurate on the 24 held-out rows where ShelveLoc = Bad.
Best R²
0.7469
Best quality
70.95 %
Best target
sales
Targets
1

Concerns

SeveritySectionConcernDetail
WarningImplicationsWeaker quality on 4 slices of the dataOn 4 of the 34 slices compared the model is weaker than on all held-out rows (R² 0.747). The largest gaps: ShelveLoc = Bad (R² 0.105 against 0.747 on all rows, on 24 rows) and Income in the highest quarter (94 to 120) (R² 0.575 against 0.747 on all rows, on 21 rows). Predictions for these rows deserve caution; the Quality by Input Range chapter lists every slice.
InfoImplicationsIf US is a characteristic predictions must not follow, Advertising stands in for itThis matters only if US is a characteristic the predictions must not follow, such as a person's age, sex or origin; for an ordinary property of the data it needs no action. Advertising and US go together with an association of 0.684 on a scale from 0 to 1. The model reads Advertising. A model that reads Advertising can tell the groups of US apart, so leaving US out of the inputs would not keep it out of the predictions. The Fairness chapter compares the model across its groups. An association, not a cause.

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.

Discovered Relations

Quality and error measured on the 80 held-out rows
TargetFormulaQualityComplexityError
SalesSales = 6.9 + 0.11 × Advertising − 0.039 × Age + 0.0934 × CompPrice − 0.0903 × Price + 3.38 × indicator(ShelveLoc = "Good")70.95 %21.9 %25.31 %

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.

Model

The model's output moves mostly with price and less with the remaining four inputs. The share of variation explained reaches 0.747 on held-out rows, and the error is 1.08. Quality is steady across the rows held out.

Most notable in the Model section:

  • Price carries most of the model (63% of its reliance on inputs)

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.

Sales

The Model section builds one model for Sales: 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.

Sales Model

The model's output relies most on Price, where it falls as Price rises, and on ShelveLoc, with higher output when ShelveLoc = Good. It also rises with CompPrice and Advertising, and falls as Age increases.

Sales = 6.9 + 0.11 × Advertising − 0.039 × Age + 0.0934 × CompPrice − 0.0903 × Price + 3.38 × indicator(ShelveLoc = "Good")f:(Advertising,Age,CompPrice,Price,ShelveLoc)↦SalesAdvertising,Age,CompPrice,Price,Sales∈RShelveLoc∈TSales=6.9+0.11×Advertising−0.039×Age+0.0934×CompPrice−0.0903×Price+3.38×ShelveLoc⩵Good
Sales = 6.9 + 0.11 × Advertising − 0.039 × Age + 0.0934 × CompPrice − 0.0903 × Price + 3.38 × indicator(ShelveLoc = "Good")
Quality 70.95 % · Error 25.31 % on 80 held-out rows · Complexity 21.9 %
Constants of the formula
ValueRoleInputUnit
6.9InterceptUnit of the model's output
0.11SlopeAdvertisingOutput units per unit of Advertising
-0.039SlopeAgeOutput units per unit of Age
0.0934SlopeCompPriceOutput units per unit of CompPrice
-0.0903SlopePriceOutput units per unit of Price
3.38Condition effectUnit of the model's output
Share of the output's variation they carry
100 %
Rows the search trained on, measured
160
Rows the search trained on
160
Held out R²
0.7469 (95 % CI 0.6576 – 0.8134)
Constant base
6.9
Constants in the formula
6
Terms that reach 95% of the output's variation
5
Terms that read an input, constants multiplied in
5
Training rows
320
Held-out rows
80
Inputs the formula reads
5
Terms in the formula
6
Inputs the formula does not read
Education, Income, Population, Urban, US

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.

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.

Prediction

The model predicts Sales 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.

Prediction Chart

R² of 0.747: good. The model captures most of the variation; individual predictions still carry uncertainty; the Prediction Interval Coverage chapter shows how wide it is. Mean absolute error is 1.08 on the target's scale.

Error
25.31 % (95 % CI 18.66 % – 34.24 %)
MAE
1.078 (95 % CI 0.9133 – 1.266)
N
80
Training rows
320
Quality
70.95 % (95 % CI 60.7 % – 78.58 %)
R²
0.7469 (95 % CI 0.6576 – 0.8134)
Target
Sales

How to read this

Actual Sales 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 colour the number of rows that fall in both, and the cells whose two ranges match hold the rows the model predicts well.

Residual Distribution
Bias
-0.0476 (95 % CI -0.3373 – 0.2336)
N
80
P50
0.09815 (95 % CI -0.2941 – 0.4286)
P90
1.888 (95 % CI 1.256 – 1.952)
P95
1.943 (95 % CI 1.696 – 2.316)
P99
2.378 (95 % CI 1.956 – 2.614)
Std
1.38 (95 % CI 1.173 – 1.588)

How to read this

How the held-out residuals (actual Sales 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.

Which inputs matter most

The model's output falls with Price on every one of the 10,240 steps between observed values. The model's output rises with CompPrice on every one of the 9,440 steps between observed values. The model's output rises with Advertising on every one of the 3,680 steps between observed values. Moving Price from 99.0 to 130, its first and third quartile, with every other input as observed lowers the model's output by 2.80 on average. These are measurements on the rows the search trained on, inside the observed ranges; outside them nothing is claimed.

InputShare of importancePermutation importanceSensitivityResidual correlationOccurrencesModel's output as the input risesChange, first to third quartile
Price62.8 %5.7730.09030.26031Falls (10,240 of 10,240 steps)-2.799 (99 to 130)
ShelveLoc16.64 %1.5291
CompPrice11.8 %1.0840.09340.26961Rises (9440 of 9440 steps)+1.868 (113 to 133)
Advertising5.01 %0.4610.11-0.13841Rises (3680 of 3680 steps)+1.21 (0 to 11)
Age3.75 %0.3450.039-0.088851Falls (8320 of 8320 steps)-0.975 (41 to 66)
Rows the search trained on, measured
160
Inputs the formula does not read
Education, Income, Population, Urban, US

How to read this

The chart ranks the inputs of the Sales 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.

Baselines

The model beats the strongest simple yardstick, a straight line through the data, by 21 points of variance explained on the held-out rows. Row by row the margin is small: on many individual rows the yardstick does about as well, and whether the complexity is justified depends on the use. Of the 80 held-out rows where the two differ, the model is closer on 65% (95% interval 54% to 75%).

Best yardstick's score
0.5407 (95 % CI 0.3823 – 0.6599)
Best baseline R²
0.5407
Effect size d
0.2942 (95 % CI 0.09685 – 0.5146)
Effect size d magnitude
small
Margin over the best yardstick
0.2061 (95 % CI 0.1005 – 0.3384)
Margin's interval stays above zero
Yes
Model MAE
1.078 (95 % CI 0.9133 – 1.266)
Model
0.7469 (95 % CI 0.6576 – 0.8134)
Model R²
0.7469 (95 % CI 0.6576 – 0.8134)
Held-out rows
80
Held-out rows without an answer
0
Target
Sales
Rows where the model is closer
65 % (95 % CI 54.08 % – 74.55 %)
Rows compared
80
Rows level
0
Regression yardstick trained with gaps filled (too few complete rows)
No
Training rows of the regression yardstick
320

How to read this

The table scores the model for Sales 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 Sales 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 Sales.

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.

Generalization

The model explains held-out rows about as well as training rows (held-out R² 0.747, gap between training and held-out R² 0.0287): it generalizes. The structure the formula encodes carries from the training rows to the held-out rows.

Generalization gap
0.02867
N test
80
Target
Sales
Test error
1.078 (95 % CI 0.9133 – 1.266)
Test R²
0.7469 (95 % CI 0.6576 – 0.8134)
Train error
0.1803
Train R²
0.7755
Verdict
Generalizing

How to read this

The chart sets the score of the Sales 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.

How each term responds to its inputs

Read as a sum, the formula has 6 additive terms: 5 that read an input and a constant base. All that read an input are needed to reach 100% of the variation of the model's output on the 160 rows the search trained on: the largest are -0.0903 × Price (43%), 3.38 × indicator(ShelveLoc = "Good") (37%) and 0.11 × Advertising (7.2%). A share describes the model's output, not the target. Each curve below shows one term's response to its free inputs, with the other inputs held at the training median.

Additive
Yes
Rows the search trained on, measured
160
Terms in the formula
6
Notation
6.9 + 0.11 × Advertising − 0.039 × Age + 0.0934 × CompPrice − 0.0903 × Price + 3.38 × indicator(ShelveLoc = "Good")
Target
Sales

How to read this

The formula for Sales 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 Sales 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.

How the formula behaves outside the data

The model diverges beyond the observed range of advertising, age, compprice, price: predictions there are not supported by the formula's shape. How the model behaves beyond the observed range of shelveloc 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
160
Inputs the formula reads
5
N singularities
0
Risky inputs
advertising, age, compprice, price
Target
Sales
Limits at neg infinity
InputKind
AdvertisingNegative infinity
AgePositive infinity
CompPriceNegative infinity
PricePositive infinity
ShelveLocUnknown
Limits at pos infinity
InputKind
AdvertisingPositive infinity
AgeNegative infinity
CompPricePositive infinity
PriceNegative infinity
ShelveLocUnknown

How to read this

The formula for Sales 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.

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 Sales. 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.

What each term is worth

Full-formula R² is 0.747. Each bar below shows how much R² drops when that term is replaced by its mean contribution. Large drop: load-bearing term; near zero: decoration the formula could lose; negative: overfit signal (ablation improves R²).

Additive
Yes
Baseline R²
0.7469
Held-out rows
80

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 Sales, 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.

What-If
The prediction at the typical row: every input at its median or most common value
ItemValue
Advertising5.5
Age55
CompPrice125
Price116
ShelveLocMedium
Term 1 adds6.9
Term 2 adds0.605
Term 3 adds-2.145
Term 4 adds11.68
Term 5 adds-10.47

The first 10 rows of 12. The other 2 are in the notebook.

Additive
Yes
Held-out rows for the interval
80
Terms in the formula
6
Residual P5
-2.593
Residual P95
1.943
Residual std
1.38

How to read this

The chapter shows the model's prediction for Sales 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.

Sensitivity

The prediction responds to 5 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
6

How to read this

For each input, the chart traces the rate at which the predicted Sales 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.

How two inputs act together

Advertising, age, compprice and price enter linearly. No term reads two inputs, so the grid below renders empty; each input's contribution can be read independently in the chapter "How each term responds to its inputs".

Additive
Yes
Inputs in different terms act separately
Yes
Rows the search trained on, measured
160
Inputs the formula is linear in
Advertising, Age, CompPrice, Price
Terms in the formula
6
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 Sales 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.

Worst Predictions

The worst held-out row was missed by 3.97 (target's scale). The 10 rows with the largest absolute residuals out of 80 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.

ActualPredictedResidualAdvertisingAgeCompPricePriceShelveLoc
3.0707.035-3.9651375118104Bad
4.1507.359-3.209338141128Bad
1.8204.870-3.050077139133Bad
3.1305.939-2.8091166130130Bad
12.1109.496+2.6141826118104Medium
7.50010.082-2.58204314091Bad
3.5805.926-2.346072142164Good
8.9706.654+2.316033132125Medium
10.5908.418+2.1721530131124Medium
10.3608.247+2.1131834107103Medium
Additive
Yes
N extrapolated
0
Held-out rows
80
Held-out rows for the interval
80
Residual P5
-2.593
Residual P95
1.943
Residual std
1.38

How to read this

The held-out rows on which the model misses Sales 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.

Prediction Interval Coverage

Across 80 held-out rows, the 95% prediction interval actually covers 95% of the rows. The normal band is too narrow for this residual distribution (worst gap 5.00 percentage points): the per-row prediction interval is a lower bound on uncertainty, and the true risk is higher than stated.

Nominal levelHalf-widthEmpirical coverage
50 %0.930652.5 %
80 %1.76875 %
90 %2.26990 %
95 %2.70495 %
Largest miss (percentage points)
5
Held-out rows
80
Residual P2.5
-3.054
Residual P97.5
2.175
Verdict
Overconfident

How to read this

Each held-out prediction of Sales 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.

Regression Diagnostics

80 held-out residuals, with a spread up to 3.97 in absolute terms. Mean residual is -0.0476, 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
-0.08005
Bias
-0.0476
Bias over residual sd
-0.0345
Held-out rows
80
Order
shuffled
Residual sd
1.38
Share within normal band
0.95

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 Sales, 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 neighbouring 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.

Search Progress

The search ran for 326 generations, ending at quality 0.725 (gain +0.725 from generation 2). The curve shows a substantial climb; if the last generations are still rising, a longer time limit is likely to produce a better model.

First generation
2
First quality
0 %
Last generation
327
Last improvement ms
2,475,110
Last quality
72.46 %
Max quality
72.46 %
N points
200
N total generations
326
Search time ms
3,600,244
Stop reason
Maximum runtime reached

How to read this

The line traces the best quality the search for the Sales 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.

Model Constraints

Across 80 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.

NameStatusDetail
Output coherentPassno non-finite predictions across 80 rows
Within domainPassno rows outside the training target's ±2σ range across 80 rows
Bounded predictionsPassno runaway predictions beyond ±10σ across 80 rows
Held-out rows
80
N non finite
0
N outside domain
0
N runaway
0
Verdict
Clean

How to read this

Sanity checks on the Sales 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.

Modeling

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

MAE of a straight line through the inputs (same rows)
1.514
RMSE of a straight line through the inputs (same rows)
1.882
Training MAE (rows the search trained on)
1.132
MAE of always predicting the mean (same rows)
2.144
RMSE of always predicting the mean (same rows)
2.629
Training R² (rows the search trained on)
0.7217
Training RMSE (rows the search trained on)
1.387
Rows the search trained on, measured
160
Error
18.03 %
Inputs the straight line used
7
N train
320
Quality
74.55 %

How to read this

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

Data

Several inputs relate strongly to either sales or each other, so their associations might make them more important for analysis of the model's steady patterns. All columns have complete data, and no leakage or duplication was found.

Most notable in the Data section:

  • ShelveLoc accounts for 31% of the spread in Sales: Good averages 4.69 above Bad (95% interval 3.98 to 5.41)
  • Advertising and US go together (association 0.684)
  • CompPrice and Price move together (correlation 0.585)

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.

Column Profiles

Profiled 11 columns across 400 rows (8 numeric, 3 non-numeric): every column is at least 75% populated. Distributional statistics on numeric columns can be read at face value.

Columns
NameRoleTypeNMissingMeanStdMinP25MedianP75MaxSkewnessCV
SalesTargetnumber40007.4962.8240.0005.3907.4909.32016.2700.18560.3767
AdvertisingInputinteger40006.6356.65000512290.63961.002
AgeInputinteger400053.3216.202539.7554.506680-0.077180.3038
CompPriceInputinteger4000125.015.377115125135175-0.042750.1227
EducationInputinteger400013.902.6210121416180.044010.1885
IncomeInputinteger400068.6627.992142.7569911200.049440.4076
PopulationInputinteger4000264.8147.410139272398.5509-0.051230.5565
PriceInputinteger4000115.823.724100117131191-0.12530.2045
ShelveLocInputtext4000
UrbanInputbool4000

The first 10 rows of 11. The other 1 is in the notebook.

N columns
11
N high missing
0
N rows
400
N strongly skewed
0
Table branch
main
Table ID
9c0fe08c-6d48-4337-9e08-d6dc836e5514

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.

Advertising

Advertising: 0 to 29, median 5, skewed to high values, 28 distinct values.

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.

Description
Advertising
StatisticValue
Typeinteger
Values present400
Distinct values28
Continuousyes
Decimal places0
Minimum0
Maximum29
Mean6.635
Median5
Mode0

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

How to read this

The figures below summarize Advertising 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 Advertising 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.

By Value
Mean per value of Advertising
AdvertisingRowsIncomeEducationCompPricePriceSalesAgePopulation
014467.3414.18124.9113.66.82553.63231.4
1855.2512.25124.6123.16.83843.38176.3
2978.3314.56125121.36.56951.78263.2
31465.3614.57128.1107.67.36357.5219.1
41268.3313.5124.71186.99351.5288.3
51972.7913132125.76.83954.53275.3
6758.1415.43121.6124.45.58450.14332.9
71664.6913128.7123.67.07846.38233.5
81169.3612.36125.5112.98.09351.45267.6
91078.812.6120.7115.78.0553.1297.2

The first 10 rows of 28. The other 18 are in the notebook.

How to read this

The rows are grouped by the value of Advertising, 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 Advertising; 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.

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
USa number and a category0.6845< 0.001400
Salestwo numbers0.27540.26950.1761 to 0.35810.27540.1948< 0.001400
Populationtwo numbers0.26570.26570.1721 to 0.35450.24190.1731< 0.001400
ShelveLoca number and a category0.05950.6286400
Incometwo numbers0.0590.059-0.03929 to 0.15610.057810.041010.5978400
Educationtwo numbers0.04622-0.03359-0.1312 to 0.06467-0.04622-0.034370.6286400
Pricetwo numbers0.044540.04454-0.05375 to 0.1420.035610.025960.6286400
Urbana number and a category0.042040.6286400
CompPricetwo numbers0.02859-0.0242-0.122 to 0.07403-0.02859-0.02050.6994400
Agetwo numbers0.005157-0.004557-0.1026 to 0.093540.0051570.0051570.9276400

How to read this

Each row pairs Advertising 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 Advertising, 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.

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 Advertising'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 Advertising (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
Visualization

The rows of Advertising 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 Advertising 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.

Age

Age: 25 to 80, median 54.5.

Description
Age
StatisticValue
Typeinteger
Values present400
Distinct values56
Continuousyes
Decimal places0
Minimum25
Maximum80
Mean53.32
Median54.5
Mode62

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

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Salestwo numbers0.2363-0.2318-0.3225 to -0.1369-0.2363-0.1592< 0.001400
Pricetwo numbers0.1184-0.1022-0.1982 to -0.004165-0.1184-0.078420.1504400
CompPricetwo numbers0.1086-0.1002-0.1964 to -0.002207-0.1086-0.075010.1504400
ShelveLoca number and a category0.057790.9276400
Populationtwo numbers0.04266-0.04266-0.1401 to 0.05562-0.04147-0.028590.9276400
Urbana number and a category0.028480.9276400
USa number and a category0.0086520.9276400
Educationtwo numbers0.0064880.006488-0.09162 to 0.10450.0044580.0029350.9276400
Advertisingtwo numbers0.005157-0.004557-0.1026 to 0.093540.0051570.0051570.9276400
Incometwo numbers0.00467-0.00467-0.1027 to 0.093430.0025940.0026170.9276400
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 -0.00 between each value and the next), so the order carries no pattern to show.

CompPrice

CompPrice: 77 to 175, median 125; moves with Price most (r 0.585).

Description
CompPrice
StatisticValue
Typeinteger
Values present400
Distinct values73
Continuousyes
Decimal places0
Minimum77
Maximum175
Mean125
Median125
Mode121

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

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Pricetwo numbers0.58480.58480.5164 to 0.64590.54180.3877< 0.001400
Populationtwo numbers0.1089-0.09471-0.191 to 0.003378-0.1089-0.073830.1948400
Agetwo numbers0.1086-0.1002-0.1964 to -0.002207-0.1086-0.075010.1948400
Incometwo numbers0.08065-0.08065-0.1773 to 0.01754-0.06603-0.04280.2681400
Salestwo numbers0.067640.06408-0.03419 to 0.16110.067640.045880.3349400
Urbana number and a category0.066590.3349400
ShelveLoca number and a category0.039170.7373400
Advertisingtwo numbers0.02859-0.0242-0.122 to 0.07403-0.02859-0.02050.7373400
Educationtwo numbers0.02520.0252-0.07304 to 0.12290.021280.014430.7373400
USa number and a category0.016870.7373400
Visualization

The rows of CompPrice are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.00 between each value and the next), so the order carries no pattern to show.

Education

Education: 10 to 18, median 14, 9 distinct values.

Description
Education
StatisticValue
Typeinteger
Values present400
Distinct values9
Continuousyes
Decimal places0
Minimum10
Maximum18
Mean13.9
Median14
Mode12

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

By Value
Mean per value of Education
EducationRowsIncomeCompPricePriceAdvertisingSalesAgePopulation
104873.54126.91167.4388.24654.17291.8
114871.52123.1111.36.8547.47155.08249.9
124967.59123.3114.76.2457.46850.96265.1
134369.98126.2121.26.3267.05251.98286.8
144061.45123.6120.37.4757.07549.6305.7
153667.94122.9112.35.7788.13256.36289.3
164769.57125.41156.5537.24455.96258.9
174968.65127116.56.6127.50352.88217.4
184066.03125.8115.26.2757.27752.98228.9
Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Populationtwo numbers0.1064-0.1064-0.2023 to -0.008412-0.1046-0.074040.3342400
USa number and a category0.078250.5908400
Incometwo numbers0.06135-0.05686-0.1541 to 0.04143-0.06135-0.043270.7499400
Salestwo numbers0.05196-0.05196-0.1492 to 0.04633-0.03447-0.022530.7499400
Advertisingtwo numbers0.04622-0.03359-0.1312 to 0.06467-0.04622-0.034370.8488400
Urbana number and a category0.033090.8488400
ShelveLoca number and a category0.029490.8971400
CompPricetwo numbers0.02520.0252-0.07304 to 0.12290.021280.014430.8791400
Pricetwo numbers0.024440.01175-0.08641 to 0.10970.024440.017140.8971400
Agetwo numbers0.0064880.006488-0.09162 to 0.10450.0044580.0029350.8971400
Visualization

The rows of Education are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.07 between each value and the next), so the order carries no pattern to show.

Income

Income: 21 to 120, median 69.

Description
Income
StatisticValue
Typeinteger
Values present400
Distinct values98
Continuousyes
Decimal places0
Minimum21
Maximum120
Mean68.66
Median69
Mode69

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

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Salestwo numbers0.15540.1520.05471 to 0.24630.15540.10310.0231400
USa number and a category0.08960.3575400
CompPricetwo numbers0.08065-0.08065-0.1773 to 0.01754-0.06603-0.04280.3575400
ShelveLoca number and a category0.072550.5012400
Educationtwo numbers0.06135-0.05686-0.1541 to 0.04143-0.06135-0.043270.4299400
Advertisingtwo numbers0.0590.059-0.03929 to 0.15610.057810.041010.4299400
Pricetwo numbers0.0567-0.0567-0.1539 to 0.04159-0.04911-0.032570.4299400
Urbana number and a category0.037970.5611400
Populationtwo numbers0.007877-0.007877-0.1058 to 0.09025-0.007395-0.005850.9258400
Agetwo numbers0.00467-0.00467-0.1027 to 0.093430.0025940.0026170.9258400
Visualization

The rows of Income are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.07 between each value and the next), so the order carries no pattern to show.

Population

Population: 10 to 509, median 272.

Description
Population
StatisticValue
Typeinteger
Values present400
Distinct values275
Continuousyes
Decimal places0
Minimum10
Maximum509
Mean264.8
Median272
Mode125

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

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Advertisingtwo numbers0.26570.26570.1721 to 0.35450.24190.1731< 0.001400
CompPricetwo numbers0.1089-0.09471-0.191 to 0.003378-0.1089-0.073830.1948400
Educationtwo numbers0.1064-0.1064-0.2023 to -0.008412-0.1046-0.074040.1671400
USa number and a category0.060560.5233400
Urbana number and a category0.052030.5233400
Salestwo numbers0.050470.05047-0.04782 to 0.14780.041940.029840.5233400
ShelveLoca number and a category0.044780.8392400
Agetwo numbers0.04266-0.04266-0.1401 to 0.05562-0.04147-0.028590.564400
Pricetwo numbers0.01214-0.01214-0.1101 to 0.08601-0.006022-0.0037740.8752400
Incometwo numbers0.007877-0.007877-0.1058 to 0.09025-0.007395-0.005850.8752400
Visualization

The rows of Population 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.

Price

Price: 24 to 191, median 117; moves with CompPrice most (r 0.585).

Description
Price
StatisticValue
Typeinteger
Values present400
Distinct values101
Continuousyes
Decimal places0
Minimum24
Maximum191
Mean115.8
Median117
Mode120

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

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
CompPricetwo numbers0.58480.58480.5164 to 0.64590.54180.3877< 0.001400
Salestwo numbers0.445-0.445-0.5203 to -0.3627-0.4076-0.2799< 0.001400
Agetwo numbers0.1184-0.1022-0.1982 to -0.004165-0.1184-0.078420.137400
USa number and a category0.057860.5158400
Incometwo numbers0.0567-0.0567-0.1539 to 0.04159-0.04911-0.032570.5158400
ShelveLoca number and a category0.05170.7349400
Urbana number and a category0.047020.5348400
Advertisingtwo numbers0.044540.04454-0.05375 to 0.1420.035610.025960.5348400
Educationtwo numbers0.024440.01175-0.08641 to 0.10970.024440.017140.8148400
Populationtwo numbers0.01214-0.01214-0.1101 to 0.08601-0.006022-0.0037740.8148400
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 0.11 between each value and the next), so the order carries no pattern to show.

Sales

Sales: 0 to 16.3, median 7.49; moves opposite to Price most (r -0.445).

Description
Sales
StatisticValue
Typenumber
Values present400
Distinct values336
Continuousyes
Decimal places2
Minimum0
Maximum16.27
Mean7.496
Median7.49
Mode7.8

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

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
ShelveLoca number and a category0.5632< 0.001400
Pricetwo numbers0.445-0.445-0.5203 to -0.3627-0.4076-0.2799< 0.001400
Advertisingtwo numbers0.27540.26950.1761 to 0.35810.27540.1948< 0.001400
Agetwo numbers0.2363-0.2318-0.3225 to -0.1369-0.2363-0.1592< 0.001400
USa number and a category0.1771< 0.001400
Incometwo numbers0.15540.1520.05471 to 0.24630.15540.10310.003849400
CompPricetwo numbers0.067640.06408-0.03419 to 0.16110.067640.045880.2871400
Educationtwo numbers0.05196-0.05196-0.1492 to 0.04633-0.03447-0.022530.3489400
Populationtwo numbers0.050470.05047-0.04782 to 0.14780.041940.029840.3489400
Urbana number and a category0.015420.7585400
Visualization

The rows of Sales 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.

ShelveLoc

ShelveLoc: 3 distinct values, most often Medium.

Description
ShelveLoc
StatisticValue
Typetext
Values present400
Distinct values3
Continuousno
ModeMedium
Rows per value
VariableValueRowsShare
ShelveLocMedium21954.75 %
ShelveLocBad9624 %
ShelveLocGood8521.25 %
By Value

ShelveLoc accounts for 31% of the spread in Sales: Good averages 4.69 above Bad (95% interval 3.98 to 5.41). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
SalesGood8510.219.674 to 10.75
SalesMedium2197.3077.005 to 7.608
SalesBad965.5235.045 to 6
Mean per value of ShelveLoc
ShelveLocRowsIncomeEducationCompPricePriceAdvertisingSalesAgePopulation
Bad9672.2413.96124114.36.2195.52352.05275.3
Good8567.9813.75125.8117.97.35310.2152.61267
Medium21967.3513.93125.1115.76.5397.30754.16259.4

Urban

Urban: 2 distinct values, most often yes.

Description
Urban
StatisticValue
Typebool
Values present400
Distinct values2
Continuousno
Modetrue
Rows per value
VariableValueRowsShare
Urbantrue28270.5 %
Urbanfalse11829.5 %
By Value
Mean per value of Urban
UrbanRowsIncomeEducationCompPricePriceAdvertisingSalesAgePopulation
false11867.0214.03123.4114.16.2037.56452.61276.7
true28269.3413.84125.6116.56.8167.46853.62259.9

US

US: 2 distinct values, most often yes.

Description
US
StatisticValue
Typebool
Values present400
Distinct values2
Continuousno
Modetrue
Rows per value
VariableValueRowsShare
UStrue25864.5 %
USfalse14235.5 %
By Value
Mean per value of US
USRowsIncomeEducationCompPricePriceAdvertisingSalesAgePopulation
false14265.2814.18124.61140.5076.82353.13252.8
true25870.5213.75125.2116.810.017.86753.43271.5

Column · ShelveLoc

ShelveLoc is profiled on its own because it may hold free text.

Column shelveloc is non-numeric, with a value in 400 rows of 400 total rows: every row carries a value, so the distribution below reflects the full population.

Column profile kind
Text categorical
N
400
N distinct shown
3
N missing
0
N rows
400
Name
shelveloc
Numeric
No
Pct missing
0
Type
text
Top distinct
CountValue
219Medium
96Bad
85Good

How to read this

A column is profiled on its own when many of its rows are missing or it may hold free text. A numeric column shows a 20-bin histogram with its mean, standard deviation and quartiles. A categorical or coded column shows its ten most frequent values. A free-text column shows a summary of its words.

Everything here describes the data as supplied. Nothing comes from the model, so the chapter reads the same whether the model is strong or weak.

Duplicate Rows

Across 400 rows the table has no exact duplicates: every row's value vector is unique across all columns, so the row count reflects distinct rows.

Largest group size
0
N exact duplicates
0
N exact groups
0
N rows
400
N unique
400
Pct exact duplicates
0
Verdict
Clean

How to read this

This chapter counts rows that repeat another row exactly, value for value: how many rows take part in a duplicate group, how many groups exist, and how large the largest group is.

Duplicate rows inflate the apparent row count and can place identical rows on both sides of the training and held-out split, which flatters every quality figure in this report. Only exact repeats are detected here; rows that differ by a rounding digit or a whitespace variant count as distinct.

Leakage Suspicion

Nothing to check. No column that awaits an answer tracks a target at 0.95 or more, so no column on its own gives a target away.

Categorical verdict
Clean
Critical threshold
0.999
N categorical critical
0
N categorical suspects
0
N confirmed
0
N critical
0
N suspects
0
Threshold
0.95
Verdict
Clean

How to read this

For every column that is not a target, this chapter reads how closely the column tracks each target, on an association from 0 to 1, and lists any pair at 0.95 or more. For two numeric columns the association is the stronger of Pearson's r and Spearman's rho; when the column or the target is categorical it is the correlation ratio η, or Cramér's V when both are. At 0.999 or more the column reproduces the target: that is what a copy or a rescaled copy of it looks like.

A listed column poses one question: is its value known before the target's value exists? If it is, the column is a strong input and the quality stated in this report is real. If it was computed from the target or recorded after it, a formula that reads it looks right on these rows and will not hold on new data. The coefficients read the same in both cases; the answer needs someone who knows where the column came from.

The owner of the analysis can give that answer. A column confirmed as known before the target is listed as confirmed and is no longer a suspect. A column that turns out to be a leak is left out of the model's inputs, and the analysis run again.

General

Clustering found no clear groups. The best grouping of the 400 complete rows over 7 numeric columns is into 2 groups, with a silhouette of 0.134 (on a scale where 1 is fully separate groups and 0 is none), below the 0.50 that clear structure shows, so no Clusters chapter is written.

Groupings tried
GroupsSilhouetteInertiaKept
20.13422381.1yes
30.12582135
40.12251953.4
50.1121875.4
60.12241739
Inertia
2381.1
K
2
N columns used
7
N rows
400
N rows complete
400
Silhouette
0.1342
Verdict
Weak

How to read this

The General chapter covers what holds across all columns together: descriptive statistics, correlations, missing values and outlier ranges under Description, the Clusters found among the rows, and, when a segment column was chosen, the Segments.

Description
Columns
VariableTypeValues presentDistinct valuesContinuousDecimal places
Incomeinteger40098yes0
ShelveLoctext4003no
Educationinteger4009yes0
Urbanbool4002no
CompPriceinteger40073yes0
Priceinteger400101yes0
Advertisinginteger40028yes0
Salesnumber400336yes2
Ageinteger40056yes0
USbool4002no

The first 10 rows of 11. The other 1 is in the notebook.

Center
VariableMeanMedianMode
Income68.666969
ShelveLocMedium
Education13.91412
Urbantrue
CompPrice125125121
Price115.8117120
Advertising6.63550
Sales7.4967.497.8
Age53.3254.562
UStrue

The first 10 rows of 11. The other 1 is in the notebook.

Spread
VariableMinimumMaximumStandard deviationVarianceInterquartile rangeSum
Income2112027.99783.248.2527,463
ShelveLoc
Education10182.6216.86745560
Urban
CompPrice7717515.33235.12049,990
Price2419123.68560.63146,318
Advertising0296.6544.23122654
Sales016.272.8247.9763.932998.5
Age258016.2262.426.2521,329
US

The first 10 rows of 11. The other 1 is in the notebook.

Shape
VariableSkewnessKurtosisTotal sum of squares
Income0.04944-1.085312,504
ShelveLoc
Education0.04401-1.2982740
Urban
CompPrice-0.042750.0416793823.8
Price-0.12530.4519223,673
Advertising0.6396-0.545117646.7
Sales0.1856-0.080883182.3
Age-0.07718-1.134104,717
US

The first 10 rows of 11. The other 1 is in the notebook.

Rows per value
VariableValueRowsShare
ShelveLocMedium21954.75 %
ShelveLocBad9624 %
ShelveLocGood8521.25 %
Urbantrue28270.5 %
Urbanfalse11829.5 %
UStrue25864.5 %
USfalse14235.5 %

How to read this

This chapter summarizes every column at once: the descriptive statistics table below, then Correlation, Missing Values and Outliers. For a numeric column the table reports its center (mean, median, mode), its spread (standard deviation, variance, interquartile range, percentiles) and its shape (skewness, kurtosis); for a categorical column it reports the count and share of each category.

Each row of the tables describes one column. The statistics are grouped: the column itself, where its values center, how far they spread, what shape their distribution takes, and its percentiles. They read as follows.

StatisticExplanation
VariableName of the column the statistics in the row describe
TypeType assigned during data processing (bool, integer, number, decimal, text or date)
Values presentNumber of rows in which the column holds a value
Distinct valuesNumber of different values the column holds
ContinuousWhether the column can take any value in a range, as opposed to a discrete column limited to a fixed set of values
Decimal placesLength of the fractional part of the values
MeanArithmetic average: the sum of all values divided by the number of values
MedianThe middle value: half of the values lie at or below it and half at or above it. It is also the 50 % percentile
ModeThe value that appears most often
MinimumSmallest numeric value in the column

The first 10 rows of 20. The other 10 are in the notebook.

Missing Values

No column has a missing value: all 400 rows are complete.

N columns with missing
0
N complete rows
400
N rows
400
Pct complete rows
100

How to read this

The table lists, for each column, the count of present values (Values) beside the row count of the processed table (Rows), the number of missing values and their share. The larger the share, the less any statistic or model built on the column can be trusted.

  • Below 5% missing: safe to use as it is.
  • 5% to 25% missing: usable; filling the gaps or excluding those rows is worth considering.
  • 25% or more missing: the column is flagged and receives its own chapter in the Data section. Verify why so much is missing before trusting any result that depends on it.

A model is computed only on rows that hold a value in every column it reads, so gaps spread over several columns cost more rows than any one column's share suggests. The last column, Complete rows without, shows for each input with gaps how many rows would be complete if that column were left out; it is blank for a column without gaps, for a target, and for a column no model reads. A column that is the only gap in many rows is the one whose removal keeps the most data.

Outliers

3 of 8 numeric columns have values beyond the Tukey fences (1.5 times the interquartile range outside the quartiles). price has the largest share: 1.3% of its values. Whether such values are rare but real rows or recording errors decides whether they are kept.

Outliers
ColumnValuesLower fenceUpper fenceBelowAboveShare
Advertising400none belownone above000 %
Age400none belownone above000 %
CompPrice40085165110.5 %
Education400none belownone above000 %
Income400none belownone above000 %
Population400none belownone above000 %
Price40053.50177.50321.25 %
Sales400none below15.2150020.5 %
N columns
8
N columns assessed
8
N columns with outliers
3
Worst column
price
Worst pct
1.25

How to read this

The table counts the outliers of every numeric column by the Tukey rule. Q1 and Q3 are the values below which 25% and 75% of a column's values fall, and the interquartile range is Q3 minus Q1. The lower fence lies 1.5 times the interquartile range below Q1, the upper fence the same distance above Q3; a value beyond a fence is an outlier. For each column the table gives both fences, how many values lie below and above them, their share of the column's values, and the single value that lies furthest beyond its fence. These are the same fences the box plots draw their whiskers to.

A column is not assessed when it has fewer than four values, or when the middle half of its values sits on a single number: the interquartile range is then zero and the rule would call every other value an outlier.

Extreme values can be real (rare but legitimate rows) or errors (measurement mistakes, data-entry slips). Which case applies decides whether to keep, clip or drop them before trusting a model that uses the column. A strongly skewed column has many values beyond one fence without any of them being wrong.

Correlation

55 column pairs analyzed across 400 rows (28 of two numbers, 24 of a number and a category, 3 of two categories); 9 (16%) are unlikely to be chance, after allowing for how many pairs were tested. Any other pair may still show a large association, but one that cannot be told apart from chance on the rows the pair was measured on. Every pair is stated as an association from 0 to 1, whatever its kind, so pairs of different kinds can be ranked against each other. An association states that two columns go together, not that one brings the other about.

N categorical pairs
3
N mixed pairs
24
N numeric pairs
28
N pairs
55
N samples
400
N significant
9
Pairs · 24 of 55 rows shown
Association95 % CIColumnEta squaredGroupsKendallKindNOther columnPearsonQ valueSpearman
0.6845Advertising0.46852numeric_categorical400US< 0.001
0.58480.5164, 0.6459CompPrice0.3877numeric400Price0.5848< 0.0010.5418
0.5632Sales0.31723numeric_categorical400ShelveLoc< 0.001
0.445-0.5203, -0.3627Price-0.2799numeric400Sales-0.445< 0.001-0.4076
0.27540.1761, 0.3581Advertising0.1948numeric400Sales0.2695< 0.0010.2754
0.26570.1721, 0.3545Advertising0.1731numeric400Population0.2657< 0.0010.2419
0.2363-0.3225, -0.1369Age-0.1592numeric400Sales-0.2318< 0.001-0.2363
0.1771Sales0.031362numeric_categorical400US0.00256
0.15540.05471, 0.2463Income0.1031numeric400Sales0.1520.014120.1554
0.1184-0.1982, -0.004165Age-0.07842numeric400Price-0.10220.2055-0.1184

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

How to read this

This chapter measures how pairs of columns go together: the Correlation Measures table below, the Strongest Correlations, the Correlation Heatmap and, when the analysis has targets, the Potential Data Leaks. Numeric columns and categorical columns are both paired, a categorical column being one that is not numeric and holds between 2 and 50 distinct values. Every pair is stated as an association from 0 (unrelated) to 1 (one column gives the other away), so pairs of different kinds can be compared. An association states that two columns go together, not that one brings the other about.

The table lists every pair of columns once, with its kind, its association, the figures of its kind, its q-value and the number of rows the pair was measured on. A pair is measured on the rows where both columns hold a value.

MeasureExplanation
KindWhat the two columns hold: two numbers (numeric), a number and a category (numeric_categorical), or two categories (categorical)
AssociationStrength of the pair from 0 to 1, whatever its kind. Two numbers: the stronger of Pearson's coefficient and Spearman's ρ, without its sign. A number and a category: the correlation ratio η. Two categories: Cramér's V
CovarianceTwo numbers only. 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 further context
Pearson's Correlation CoefficientTwo numbers only. Strength and direction of a linear relationship, from -1 (perfect negative) through 0 (no linear relationship) to 1 (perfect positive)
Spearman's ρTwo numbers only. Strength and direction of a monotonic relationship, computed on ranks rather than values. It captures relationships that consistently rise or fall even when they are not linear, and it is less affected by extreme values
Kendall's τTwo numbers only. Strength and direction of a monotonic relationship based on the concordance of ranks between the two columns. More conservative than Spearman's ρ, and steadier on small tables
Correlation ratio ηA number and a category. The square root of the share of the number's spread that lies between the categories, from 0 (every category has the same average) to 1 (the category fixes the number). For a category of two values it equals the absolute Pearson coefficient with the category coded 0 and 1. Its square, η², is that share: an η of 0.6 means the categories account for 36% of the spread
Cramér's VTwo categories. How far the table of counts of the two columns departs from what unrelated columns would give, from 0 (unrelated) to 1 (each value of one column goes with one value of the other), corrected so that columns with many rare values do not read as related
q-valueHow likely a pair this strong is among unrelated columns, adjusted for the number of pairs listed (Benjamini-Hochberg false-discovery rate). Among all pairs reported with q below 0.05, about 5% or fewer are expected to be false discoveries. The right measure for ranking many pairs
95% CI (Pearson)Two numbers only. Lower and upper bounds of the 95% confidence interval on the Pearson coefficient, via the Fisher z-transform, on the rows the pair was measured on. A narrow interval means the coefficient is well estimated; a wide interval that straddles zero means the relationship is uncertain

The first 10 rows of 12. The other 2 are in the notebook.

Strongest Correlations

The 20 pairs of columns with the smallest q-values: the associations that remain significant after the Benjamini-Hochberg adjustment for the number of pairs tested. Among all pairs reported with q below 0.05, about 5% or fewer are expected to be false discoveries; a pair above 0.05 may be coincidence even when its association looks strong.

A pair with an association above 0.9 often signals a duplicate or derived column. Moderately strong pairs with a low q-value are the relationships most worth investigating.

Each listed pair is drawn below, the picture behind the figure. Two numbers are drawn as a scatter, one point per row, the quickest way to see whether a strong value rests on a trend or on a few rows far from the rest. A number and a category are drawn as the number's box plot per category. Two categories are drawn as a heat map of how many rows hold each combination of values.

Potential Data Leaks

The table lists the association between every column and each target, whatever the two hold, with its q-value. An association close to 1 with a target, together with a q-value near zero, is one of two things: a strong input, or a data leak, a column derived from the target after the fact, such as a tax amount computed from the price being predicted. A leaked column trains a model that looks perfect and fails on new data, because the leak is not available at prediction time.

The Leakage Suspicion chapter in the Data section lists the pairs at 0.95 or more and marks those at 0.999 or more, where a column reproduces the target. For two numbers the association is whichever of Pearson's r and Spearman's rho is the stronger, because a column derived from the target often follows a curve rather than a line: the log of a price tracks the price perfectly, and Pearson alone reads that as weaker than it is. For a number and a category it is the correlation ratio η, and for two categories Cramér's V. Whether a listed column is a strong input or a leak depends on whether it is known before the target exists, which the owner of the analysis can confirm.

Implications

Lower accuracy for predictions where income or advertising are high means outcomes may be less reliable for those segments. Concerns were flagged for quality and advertising as a stand-in for US; see the Concerns chapter.

Most notable in the Implications section:

  • The model is less accurate on rows with Income in the highest quarter (94 to 120) (17 points less variance explained than overall)
  • The model is less accurate on rows with Advertising in the highest quarter (12 to 29) (13 points less variance explained than overall)

How to read this

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.

Stakeholder Summary

The Sales model uses 5 of the 10 available inputs and relies most heavily on Price: as Price rises across its observed range, the model’s output falls. For this input, the analysis finds that moving Price from 99.0 to 130 lowers the model’s output by about 2.80 on average, with every other input as observed. The model’s answer stands out from simpler approaches by consistently adding a modest improvement over what a straight line through the data would capture, being closer to the true values on nearly two-thirds of the rows tested.

The model leaves out half the available columns and, while it tracks most rows well, it predicts Sales much less accurately when ShelveLoc is Bad. It also performs less strongly for high Advertising, high Income, and non-US rows. The model’s formula is short but uses specialist functions, so interpretation for decision-making may require technical support. Before operational use, validate its performance on a new sample, with special attention to areas where accuracy was lower.

Recommended Next Data

  • Collect more rows where ShelveLoc = Bad. Reason: the model is far less accurate on the 24 held-out rows in this segment, explaining 11% of the variation there.
  • Collect more rows where Income is in the highest quarter (94 to 120), Advertising is in the highest quarter (12 to 29), and US = no. Reason: the model is less accurate than overall on these segments, explaining 58% to 64% of the variation on 21 to 35 held-out rows in each.
  • Inspect predictions for ShelveLoc for parity and bias. Reason: a fairness flag was raised for the column ShelveLoc with the verdict parity and bias.

Quality by Input Range

Quality by Input Range splits the held-out rows by every usable input and compares the model's quality on each segment with its quality on all held-out rows. 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 input is split into quartiles; a categorical input 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.
Sales by Input Range

Segment shelveloc = Bad is much weaker than the aggregate (24 held-out rows): segment R² 0.105 against aggregate R² 0.747 (gap -0.642). A drop this large means the model describes this slice of the data poorly; an input interaction or a missing column may be specific to this subset.

SegmentRowsR²Gap to all rowsVerdict
ShelveLoc = Bad240.1052-0.64167Much weaker than on all rows
Income = q4210.5752-0.17167Weaker than on all rows
Advertising = q4230.6176-0.12930Weaker than on all rows
US = false350.6436-0.10329Weaker than on all rows
Price = q1210.6508-0.09607In line with all rows
Age = q4220.6810-0.06583In line with all rows
Advertising = q2430.6826-0.06424In line with all rows
Population = q2210.7038-0.04307In line with all rows
Price = q3220.7068-0.04012In line with all rows
Education = q2230.7363-0.01056In line with all rows

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

Aggregate R²
0.7469
Measure
r2
Target
Sales

Rows unlike the ones the model learned from

4 of 80 held-out rows (5%) lie further from the training rows than 99 in 100 rows like them would, measured on the 4 inputs the model reads (advertising, age, compprice, price). That is a small minority; the held-out figures describe the model on rows like the ones it was trained on.

TargetInputs measuredLineHeld-out rowsBeyond the lineShare
SalesAdvertising, Age, CompPrice, Price3.6448045 %
Score bins · 24 of 30 rows shown
CountFromTo
000.175
00.1750.35
10.350.525
10.5250.7
20.70.875
30.8751.05
21.051.225
41.2251.4
71.41.575
111.5751.75

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.

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.

Sales Fairness

Column ShelveLoc splits the held-out rows into 3 groups. The prediction-parity ratio is 0.733 (smallest over largest mean prediction), the largest absolute group residual mean is 1.28 (0.929 times the overall residual standard deviation). Parity gap and residual bias: the model both predicts systematically lower for some groups and errs systematically against at least one group. This is the worst-case fairness signal on this column; review before deployment.

Aggregate residual std
1.38
N columns
3
N proxies
1
Proxy threshold
0.5
Columns
Bias max abs residual meanColumnGroupsInterpretationN groupsN proxiesPrediction parity ratioVerdict
1.282ShelveLocMedium, Good, BadColumn ShelveLoc splits the held-out rows into 3 groups. The prediction-parity ratio is 0.733 (smallest over largest mean prediction), the largest absolute group residual mean is 1.28 (0.929 times the overall residual standard deviation). Parity gap and residual bias: the model both predicts systematically lower for some groups and errs systematically against at least one group. This is the worst-case fairness signal on this column; review before deployment.30.7334Parity and bias
0.161Urbanfalse, trueColumn Urban splits the held-out rows into 2 groups. The prediction-parity ratio is 0.983 (smallest over largest mean prediction), the largest absolute group residual mean is 0.161 (0.117 times the overall residual standard deviation). Balanced: every group is within the 80% parity band and no group's residual mean exceeds 30% of the overall residual standard deviation. No systematic disparity was detected on this column.20.9828Balanced
0.1476UStrue, falseColumn US splits the held-out rows into 2 groups. The prediction-parity ratio is 0.926 (smallest over largest mean prediction), the largest absolute group residual mean is 0.148 (0.107 times the overall residual standard deviation). Balanced: every group is within the 80% parity band and no group's residual mean exceeds 30% of the overall residual standard deviation. No systematic disparity was detected on this column. advertising goes closely with us (association 0.684 on a scale from 0 to 1): a model that reads advertising can tell the groups of us apart, so leaving us out of the inputs does not keep it out of the predictions.210.9262Balanced

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.

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.

Deployment Readiness

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

  • Targets: 1 target: sales
  • Rows: 400 rows
  • Per-Segment Predictions: not produced, because no segment column was chosen
  • Fairness: compared across 3 categorical columns (shelveloc, urban, us); 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 (4 pass, 0 warn, 2 fail), at least one check fails or a critical concern is open: Not ready. The failing checks need addressing first; they typically indicate data leakage, severe miscalibration, or a model that loses to its baseline.

NameStatusDetail
GeneralizationPassheld-out quality holds up against training quality
Prediction Interval CoverageFailintervals miss more rows than their level allows, largest gap 5.00 percentage points
Baseline Lift for SalesPassbeats the best baseline, R² lift 0.206
Model ConstraintsPassevery held-out prediction is finite and within the training range
FairnessFail3 columns compared, 1 flagged: ShelveLoc (a group both receives smaller predictions and is systematically mis-predicted)
Out-of-DistributionPass5% of held-out rows are out of distribution
N critical concerns
0
N fail
2
N pass
4
N warn
0
Overall verdict
Not ready

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.

ItemValue
Analysis IDf1ec5b50-48b7-11f1-8087-d378640fe787
Run time2026-10-23 23:03 UTC
Total rows400
Modeled targetsSales
Segment columnnot available
Time limit1 h 0 min
Time taken1 h 1 min
Credit budget500 credits
Ufinq version0.3.43

Findings

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

Loading Diafunc…