Analysis report

Auto Insurance Price

Data from UCI Machine Learning Repository, published under https://creativecommons.org/licenses/by/4.0/.

Target
Target
Analyzed
2026-10-24 23:53 UTC

Summary

The model for Target in this analysis uses only Engine size, and its output rises as Engine size increases. While the model accounts for a large share of the variation in Target and tracks Engine size closely, it falls just short of matching the predictive strength of a straight line through the data, mirroring its performance in most cases but not offering a consistent improvement. The model's answer for a typical move in Engine size, such as going from 98.0 to 141, is to raise the model's output by about 5,962 on average, with everything else as observed.

As the model does not outperform a straight line and was closer than that benchmark on less than half the held-out rows, its results should be seen as tracking the main trend in the data rather than sharpening predictions for individual cases. Accuracy also drops noticeably for higher values of Target and for the rows where Drive wheels is in the third quarter, indicating less reliable predictions in those segments. Before acting on this model, prioritize checking its predictions on new or high-magnitude cases and review performance in segments with lower accuracy.

Formula

Target = 55 × Engine size + 0.35 × Engine size^2Target=55×Engine size+0.35×Engine size2
Target = 55 × Engine size + 0.35 × Engine size^2
2 terms · additively decomposable

Held-out diagnostics

TargetQualityErrorR²MAEGeneralizationRows
Target80.13 %17.19 %0.85072547generalizing40
  • Target · Quality 95 % CI 53.99 % – 90.2 %
  • Target · Error 95 % CI 8.48 % – 39.82 %
  • Target · R² 95 % CI 0.667 – 0.9285
  • Target · MAE 95 % CI 1790.9 – 3303.8

Concerns

WARNING model

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

Across 40 held-out rows split into thirds by predicted value, the residual standard deviation in the top third is 5,410, against 2,987 in the bottom third, 1.81 times as large. The bottom third holds predictions up to 8,751 and the top third predictions from 15,491. The residual spread is 1.81 times as large at the high end: noticeable, not severe. Predictions at the high end deserve caution; a log transform of the target may tighten them.

WARNING model

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

The model explains 85% of the variation on the held-out rows, where a straight line through the data explains 88%: 2.6 points less. The interval of the margin includes zero: on this many held-out rows the two are not told apart. The formula adds little over the reference predictor; whether it is worth its complexity depends on the use. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had. Row by row the two are about level.

WARNING implications

Much weaker quality on rows with `Drive wheels` in the third quarter (1)

R² on this segment is 0.479 below the overall value, on 23 rows. The model is noticeably weaker on these rows; predictions for them deserve caution.

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 2026-10-24 23:54 UTC, analysis ed1557d0-48b7-11f1-8087-d378640fe787, 1 target, time budget 1 h 0 min, compute budget 500 credits, 201 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

Target rises with Engine size.

  1. Engine size is the only input the formula reads. Everything the formula explains about Target comes from Engine size.
  2. Target rises with Engine size. With the other inputs at typical values, the formula's value rises by 41,500 between Engine size 61 and 293. The change is not even across the range.
  3. Curb weight is not in the formula. It has the strongest correlation with Target in the table (0.83). Rows with more of it have more Engine size (correlation 0.85), and Engine size is in the formula.
  4. The formula does worse than a straight line. The formula explains 85% of Target; a straight line explains more, 88%. That was measured on 40 rows it never saw. The formula typically misses Target by 2550. It is weakest where Drive wheels is 1, where it explains 37%.

Summary

What to do next: Intervals are wider than their level requires, largest gap 10.00 percentage points.

5 chapters, each headed by what it found.

Engine size carries the whole formula

What the formula relies on. Target rises as Engine size rises.

The model's output rises with Engine size on every one of the 5,282 steps between observed values. Moving Engine size from 98.0 to 141, its first and third quartile, with every other input as observed raises the model's output by 5,962 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
Engine size100 %5.818139.7-0.52211Rises (5282 of 5282 steps)+5962 (98 to 141)
Rows the search trained on, measured
139
Inputs the formula does not read
Aspiration, Body style, Bore, City mpg, Compression ratio, Curb weight, Drive wheels, Engine location, Engine type, Fuel system, Fuel type, Height, Highway mpg, Horsepower, Length, Make, Number of engine cylinders, Number of doors, Peak rpm, Stroke, Wheel base, Width

Read as a sum, the formula has 2 additive terms. All of them are needed to reach 100% of the variation of the model's output on the 139 rows the search trained on: the largest are 0.35 × Engine size^2 (69%) and 55 × Engine size (31%). 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
139
Terms in the formula
2
Notation
55 × Engine size + 0.35 × Engine size^2
Target
Target

A straight line does better here: R² 0.85 against 0.88

How well it holds. The formula explains 85% of Target; a straight line explains more, 88%.

R² of 0.851: 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 2,547 on the target's scale.

Error
17.19 % (95 % CI 8.48 % – 39.82 %)
MAE
2547 (95 % CI 1790.9 – 3303.8)
N
40
Training rows
154
Quality
80.13 % (95 % CI 53.99 % – 90.2 %)
R²
0.8507 (95 % CI 0.667 – 0.9285)
Target
Target

The model does not beat the simple yardsticks: a straight line through the data is ahead by 2.6 points of variance explained on the held-out rows. The model's complexity is not buying anything yet. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had. Of the 40 held-out rows where the two differ, the model is closer on 45% (95% interval 31% to 60%). The interval of the margin includes zero: on this many held-out rows the two are not told apart.

Best yardstick's score
0.8771 (95 % CI 0.7588 – 0.9138)
Best baseline R²
0.8771
Effect size d
-0.0504 (95 % CI -0.3864 – 0.2463)
Effect size d magnitude
negligible
Margin over the best yardstick
-0.02635 (95 % CI -0.1247 – 0.03351)
Margin's interval stays above zero
No
Model MAE
2547 (95 % CI 1790.9 – 3303.8)
Model
0.8507 (95 % CI 0.667 – 0.9285)
Model R²
0.8507 (95 % CI 0.667 – 0.9285)
Held-out rows
40
Held-out rows without an answer
0
Target
Target
Rows where the model is closer
45 % (95 % CI 30.71 % – 60.17 %)
Rows compared
40
Rows level
0
Regression yardstick trained with gaps filled (too few complete rows)
No
Held-out rows with a gap filled for the regression yardstick
1
Training rows of the regression yardstick
156

The model explains held-out rows about as well as training rows (held-out R² 0.851, gap between training and held-out R² -0.156): it generalizes. The structure the formula encodes carries from the training rows to the held-out rows.

Generalization gap
-0.1558
N test
40
Target
Target
Test error
2547 (95 % CI 1790.9 – 3303.8)
Test R²
0.8507 (95 % CI 0.667 – 0.9285)
Train error
0.2533
Train R²
0.6949
Verdict
Generalizing

Weaker in the third quarter of Drive wheels

Where it is weak. The largest miss on a held-out row is 10,100 (recorded 36,880, predicted 26,800). 5% of the held-out rows are unlike the rows it learned from.

Segment drive_wheels in the third quarter (1) is much weaker than the aggregate (23 held-out rows): segment R² 0.372 against aggregate R² 0.851 (gap -0.479). 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
Drive wheels = q3230.3719-0.479Much weaker than on all rows
Compression ratio = q4200.7530-0.098In line with all rows
Fuel system = q2220.7914-0.059In line with all rows
Engine type = q4360.8261-0.025In line with all rows
Number of doors = true210.8327-0.018In line with all rows
Engine location = false400.8508+0.000In line with all rows
Fuel type = true340.8621+0.011In line with all rows
Number of engine cylinders = q4390.8743+0.024In line with all rows
Aspiration = false340.8752+0.024In line with all rows
Highway mpg = q38-6.7179-7.569Too few rows to judge

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

Aggregate R²
0.8508
Measure
r2
Target
Target

The worst held-out row was missed by 10,097 (target's scale). The 10 rows with the largest absolute residuals out of 40 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.

ActualPredictedResidualEngine size
36,88026,783+10,097209
31,60021,786+9814183
15,6456640+900580
844915,491-7042146
12,62917,098-4469156
11,19915,491-4292146
12,96417,098-4134156
25,55221,786+3766183
45,40049,066-3666304
51188628-351097
Additive
Yes
N extrapolated
0
Held-out rows
40
Held-out rows for the interval
40
Residual P5
-4300.4
Residual P95
9045.4
Residual std
3614.3

2 of 40 held-out rows (5%) lie further from the training rows than 99 in 100 rows like them would, measured on the 1 input the model reads (engine_size). 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
TargetEngine size2.5764025 %
Score bins · 24 of 30 rows shown
CountFromTo
500.149
10.1490.297
60.2970.446
20.4460.595
90.5950.744
80.7440.892
20.8921.041
11.0411.19
01.191.339
31.3391.487

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

Usable with care: 3 points to watch

What to watch. First: Residual variance grows with predicted value: model accuracy degrades on higher-magnitude predictions.

SeveritySectionConcernDetail
WarningModelResidual variance grows with predicted value: model accuracy degrades on higher-magnitude predictionsAcross 40 held-out rows split into thirds by predicted value, the residual standard deviation in the top third is 5,410, against 2,987 in the bottom third, 1.81 times as large. The bottom third holds predictions up to 8,751 and the top third predictions from 15,491. The residual spread is 1.81 times as large at the high end: noticeable, not severe. Predictions at the high end deserve caution; a log transform of the target may tighten them.
WarningModelThe model does not beat a straight line through the data on TargetThe model explains 85% of the variation on the held-out rows, where a straight line through the data explains 88%: 2.6 points less. The interval of the margin includes zero: on this many held-out rows the two are not told apart. The formula adds little over the reference predictor; whether it is worth its complexity depends on the use. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had. Row by row the two are about level.
WarningImplicationsMuch weaker quality on rows with Drive wheels in the third quarter (1)R² on this segment is 0.479 below the overall value, on 23 rows. The model is noticeably weaker on these rows; predictions for them deserve caution.

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

  • Targets: 1 target: target
  • Rows: 201 rows
  • Per-Segment Predictions: not produced, because no segment column was chosen
  • Fairness: compared across 4 categorical columns (aspiration, engine_location, fuel_type, number_of_doors); which of them are protected attributes is not decided here
  • Out-of-Distribution: held-out rows are scored against the training rows; predictions made outside this report are not scored
  • Provenance & Audit: recorded in the chapter of that name

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

NameStatusDetail
GeneralizationPassheld-out quality holds up against training quality
Prediction Interval CoverageWarnintervals are wider than their level requires, largest gap 10.00 percentage points
Baseline Lift for TargetWarnis not told apart from the best baseline, R² lift -0.0264
Model ConstraintsPassevery held-out prediction is finite and within the training range
FairnessWarn3 columns compared, 3 flagged: Aspiration (a group receives systematically smaller predictions), Fuel type (a group is systematically over- or under-predicted), Number of doors (a group is systematically over- or under-predicted)
Out-of-DistributionPass5% of held-out rows are out of distribution
N critical concerns
0
N fail
0
N pass
3
N warn
3
Overall verdict
Caution

Column Aspiration splits the held-out rows into 2 groups. The prediction-parity ratio is 0.685 (smallest over largest mean prediction), the largest absolute group residual mean is 647 (0.179 times the overall residual standard deviation). Parity gap: at least one group receives systematically smaller predictions on average. Whether that is a property of the target itself or a model bias against the group needs checking.

Aggregate residual std
3614.3
N columns
3
N proxies
2
Proxy threshold
0.5
Columns
Bias max abs residual meanColumnGroupsInterpretationN groupsN proxiesPrediction parity ratioVerdict
647.3Aspirationtrue, falseColumn Aspiration splits the held-out rows into 2 groups. The prediction-parity ratio is 0.685 (smallest over largest mean prediction), the largest absolute group residual mean is 647 (0.179 times the overall residual standard deviation). Parity gap: at least one group receives systematically smaller predictions on average. Whether that is a property of the target itself or a model bias against the group needs checking.20.6854Prediction parity gap
1704.2Fuel typefalse, trueColumn Fuel type splits the held-out rows into 2 groups. The prediction-parity ratio is 0.962 (smallest over largest mean prediction), the largest absolute group residual mean is 1,704 (0.472 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. compression_ratio goes closely with fuel_type (association 0.985 on a scale from 0 to 1): a model that reads compression_ratio can tell the groups of fuel_type apart, so leaving fuel_type out of the inputs does not keep it out of the predictions.210.9618Residual bias
1130.1Number of doorsfalse, trueColumn Number of doors splits the held-out rows into 2 groups. The prediction-parity ratio is 0.931 (smallest over largest mean prediction), the largest absolute group residual mean is 1,130 (0.313 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. height goes closely with number_of_doors (association 0.546 on a scale from 0 to 1): a model that reads height can tell the groups of number_of_doors apart, so leaving number_of_doors out of the inputs does not keep it out of the predictions.210.9312Residual bias

More rows in the third quarter of Drive wheels would help most

What to collect next. Its R² there is 0.37 against 0.85 on all rows, on 23 held-out rows.

  • Collect more rows where Drive wheels is in the third quarter (1). Reason: the model is far less accurate on the 23 held-out rows where Drive wheels is in the third quarter (1), explaining 37% of the variation there.
  • Add columns that record details about aspiration, fuel type, and number of doors. Reason: fairness diagnostics found a prediction parity gap for aspiration and that the model's errors differ between groups for fuel type and number of doors.
  • Collect more rows with higher-magnitude values of the model's output. Reason: model accuracy degrades on higher-magnitude predictions, as residual variance grows with predicted value.
  • Add columns that record other characteristics of the data. Reason: the model does not beat a straight line through the data on Target, indicating that further predictors could increase model quality.

Evidence

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

Headline

Diafunc modeled Target from 23 columns, using only Engine size. The model explains 85% of Target and relies most on Engine size, with which the model's output rises.

  • One target was modeled; 40 rows were held out for checking.
  • A straight line through the data already explained 88%; the model did not beat it, falling behind by 2.6 points in variance explained, and was closer than the yardstick on 45% of the held-out rows.
  • The model is far less accurate on the 23 held-out rows where Drive wheels is in the third quarter (1).
Best R²
0.8508
Best quality
80.13 %
Best target
target
Targets
1

Concerns

SeveritySectionConcernDetail
WarningModelResidual variance grows with predicted value: model accuracy degrades on higher-magnitude predictionsAcross 40 held-out rows split into thirds by predicted value, the residual standard deviation in the top third is 5,410, against 2,987 in the bottom third, 1.81 times as large. The bottom third holds predictions up to 8,751 and the top third predictions from 15,491. The residual spread is 1.81 times as large at the high end: noticeable, not severe. Predictions at the high end deserve caution; a log transform of the target may tighten them.
WarningModelThe model does not beat a straight line through the data on TargetThe model explains 85% of the variation on the held-out rows, where a straight line through the data explains 88%: 2.6 points less. The interval of the margin includes zero: on this many held-out rows the two are not told apart. The formula adds little over the reference predictor; whether it is worth its complexity depends on the use. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had. Row by row the two are about level.
WarningImplicationsMuch weaker quality on rows with Drive wheels in the third quarter (1)R² on this segment is 0.479 below the overall value, on 23 rows. The model is noticeably weaker on these rows; predictions for them deserve caution.

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 40 held-out rows
TargetFormulaQualityComplexityError
TargetTarget = 55 × Engine size + 0.35 × Engine size^280.13 %8.8 %17.19 %

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 formula includes only engine_size, with 55 × engine_size + 0.35 × engine_size^2 explaining the model's behavior. Two issues were flagged: degrading accuracy for higher values and no improvement over a straight line through target, as detailed in the Concerns chapter.

Most notable in the Model section:

  • Engine size carries most of the model (100% 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.

Target

The Model section builds one model for Target: 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.

Target Model

The model's output rises with the input Engine size, which accounts for 100% of the variation explained by the formula Target = 55 × Engine size + 0.35 × Engine size^2.

Target = 55 × Engine size + 0.35 × Engine size^2f:Engine size↦TargetEngine size,Target∈RTarget=55×Engine size+0.35×Engine size2
Target = 55 × Engine size + 0.35 × Engine size^2
Quality 80.13 % · Error 17.19 % on 40 held-out rows · Complexity 8.8 %
Constants of the formula
ValueRoleInputUnit
55SlopeEngine sizeOutput units per unit of Engine size
0.35Coefficient
Share of the output's variation they carry
100 %
Rows the search trained on, measured
139
Rows the search trained on
139
Held out R²
0.8507 (95 % CI 0.667 – 0.9285)
Constants in the formula
2
Terms that reach 95% of the output's variation
2
Training rows
154
Held-out rows
40
Inputs the formula reads
1
Terms in the formula
2
Inputs the formula does not read
Aspiration, Body style, Bore, City mpg, Compression ratio, Curb weight, Drive wheels, Engine location, Engine type, Fuel system, Fuel type, Height, Highway mpg, Horsepower, Length, Make, Number of engine cylinders, Number of doors, Peak rpm, Stroke, Wheel base, Width

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 Target 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 0.851: 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 2,547 on the target's scale.

Error
17.19 % (95 % CI 8.48 % – 39.82 %)
MAE
2547 (95 % CI 1790.9 – 3303.8)
N
40
Training rows
154
Quality
80.13 % (95 % CI 53.99 % – 90.2 %)
R²
0.8507 (95 % CI 0.667 – 0.9285)
Target
Target

How to read this

Actual Target 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
-253.9 (95 % CI -1330.2 – 880.9)
N
40
P50
-653.8 (95 % CI -1747.8 – -184.4)
P90
3311.2 (95 % CI 695.4 – 9813.8)
P95
9045.4 (95 % CI 2543.2 – 10096.7)
P99
9986.4 (95 % CI 3337.7 – 10096.7)
Std
3614.3 (95 % CI 2206.9 – 4544)

How to read this

How the held-out residuals (actual Target 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 rises with Engine size on every one of the 5,282 steps between observed values. Moving Engine size from 98.0 to 141, its first and third quartile, with every other input as observed raises the model's output by 5,962 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
Engine size100 %5.818139.7-0.52211Rises (5282 of 5282 steps)+5962 (98 to 141)
Rows the search trained on, measured
139
Inputs the formula does not read
Aspiration, Body style, Bore, City mpg, Compression ratio, Curb weight, Drive wheels, Engine location, Engine type, Fuel system, Fuel type, Height, Highway mpg, Horsepower, Length, Make, Number of engine cylinders, Number of doors, Peak rpm, Stroke, Wheel base, Width

How to read this

The chart ranks the inputs of the Target 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 does not beat the simple yardsticks: a straight line through the data is ahead by 2.6 points of variance explained on the held-out rows. The model's complexity is not buying anything yet. A plain regression on the same inputs does as well or better, so the inputs do carry signal; the search did not find a better formula in the time it had. Of the 40 held-out rows where the two differ, the model is closer on 45% (95% interval 31% to 60%). The interval of the margin includes zero: on this many held-out rows the two are not told apart.

Best yardstick's score
0.8771 (95 % CI 0.7588 – 0.9138)
Best baseline R²
0.8771
Effect size d
-0.0504 (95 % CI -0.3864 – 0.2463)
Effect size d magnitude
negligible
Margin over the best yardstick
-0.02635 (95 % CI -0.1247 – 0.03351)
Margin's interval stays above zero
No
Model MAE
2547 (95 % CI 1790.9 – 3303.8)
Model
0.8507 (95 % CI 0.667 – 0.9285)
Model R²
0.8507 (95 % CI 0.667 – 0.9285)
Held-out rows
40
Held-out rows without an answer
0
Target
Target
Rows where the model is closer
45 % (95 % CI 30.71 % – 60.17 %)
Rows compared
40
Rows level
0
Regression yardstick trained with gaps filled (too few complete rows)
No
Held-out rows with a gap filled for the regression yardstick
1
Training rows of the regression yardstick
156

How to read this

The table scores the model for Target 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 Target 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 Target.

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 0.851, gap between training and held-out R² -0.156): it generalizes. The structure the formula encodes carries from the training rows to the held-out rows.

Generalization gap
-0.1558
N test
40
Target
Target
Test error
2547 (95 % CI 1790.9 – 3303.8)
Test R²
0.8507 (95 % CI 0.667 – 0.9285)
Train error
0.2533
Train R²
0.6949
Verdict
Generalizing

How to read this

The chart sets the score of the Target 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 2 additive terms. All of them are needed to reach 100% of the variation of the model's output on the 139 rows the search trained on: the largest are 0.35 × Engine size^2 (69%) and 55 × Engine size (31%). 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
139
Terms in the formula
2
Notation
55 × Engine size + 0.35 × Engine size^2
Target
Target

How to read this

The formula for Target 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 Target 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 engine_size: predictions there are not supported by the formula's shape. Extrapolate with caution.

Bounded at infinity
No
Rows the search trained on, measured
139
Inputs the formula reads
1
N singularities
0
Risky inputs
engine_size
Target
Target
Limits at neg infinity
InputKind
Engine sizeUnknown
Limits at pos infinity
InputKind
Engine sizePositive infinity

How to read this

The formula for Target 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 Target. 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 0.851. 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
0.8507
Held-out rows
40

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 Target, 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
Engine size119
Term 1 adds6545
Term 2 adds4956.4
Prediction11501.3
Additive
Yes
Held-out rows for the interval
40
Terms in the formula
2
Residual P5
-4300.4
Residual P95
9045.4
Residual std
3614.3

How to read this

The chapter shows the model's prediction for Target 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 1 input. The formula is additive: each curve below shows the change in the prediction per unit of one input (∂ŷ/∂xᵢ) independently. Flat curves mean a constant linear response, bent curves a non-linear response that depends on the position in the range, and zero crossings a response that reverses sign between regions.

Additive
Yes
Terms in the formula
2

How to read this

For each input, the chart traces the rate at which the predicted Target 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

The formula reads one input, so there is no pair of inputs that could act together. How the model's output moves with that input is 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
139
Terms in the formula
2
Input pairs not tested
0

How to read this

When a single term of the formula combines two or more inputs, the prediction's response to one of them depends on the value of the other. This chapter draws one surface for each such pair: the predicted Target 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 10,097 (target's scale). The 10 rows with the largest absolute residuals out of 40 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.

ActualPredictedResidualEngine size
36,88026,783+10,097209
31,60021,786+9814183
15,6456640+900580
844915,491-7042146
12,62917,098-4469156
11,19915,491-4292146
12,96417,098-4134156
25,55221,786+3766183
45,40049,066-3666304
51188628-351097
Additive
Yes
N extrapolated
0
Held-out rows
40
Held-out rows for the interval
40
Residual P5
-4300.4
Residual P95
9045.4
Residual std
3614.3

How to read this

The held-out rows on which the model misses Target 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 40 held-out rows, the 95% prediction interval actually covers 92.5% of the rows. The normal band is wider than needed for this residual distribution (worst gap 10.00 percentage points), often because a few large residuals inflate σ above what the bulk requires. Real uncertainty for typical rows is tighter than the interval suggests.

Nominal levelHalf-widthEmpirical coverage
50 %2437.860 %
80 %463290 %
90 %5945.190 %
95 %708492.5 %
Largest miss (percentage points)
10
Held-out rows
40
Residual P2.5
-4532.9
Residual P97.5
9820.9
Verdict
Underconfident

How to read this

Each held-out prediction of Target 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

40 held-out residuals, with a spread up to 10,097 in absolute terms. Mean residual is -254, 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
0.3384
Bias
-253.9
Bias over residual sd
-0.07024
Held-out rows
40
Order
shuffled
Residual sd
3614.3
Share within normal band
0.925

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 Target, 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 23 generations, ending at quality 0.732 (gain +0.0456 from generation 2). The curve climbs modestly: most of the lift happened early, so the search has largely converged. A slightly longer time limit may extract a few more points.

First generation
2
First quality
68.66 %
Last generation
24
Last improvement ms
2,742,621
Last quality
73.22 %
Max quality
73.22 %
N points
23
N total generations
23
Search time ms
4,500,249
Stop reason
Maximum runtime reached

How to read this

The line traces the best quality the search for the Target 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 40 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 40 rows
Within domainPassno rows outside the training target's ±2σ range across 40 rows
Bounded predictionsPassno runaway predictions beyond ±10σ across 40 rows
Held-out rows
40
N non finite
0
N outside domain
0
N runaway
0
Verdict
Clean

How to read this

Sanity checks on the Target 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)
2169.6
RMSE of a straight line through the inputs (same rows)
2966
Training MAE (rows the search trained on)
3103.3
MAE of always predicting the mean (same rows)
5905.4
RMSE of always predicting the mean (same rows)
7850.8
Training R² (rows the search trained on)
0.7042
Training RMSE (rows the search trained on)
4269.6
Rows the search trained on, measured
139
Error
25.33 %
Inputs the straight line used
19
N train
161
Quality
70.27 %

How to read this

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

Findings

2 findings from the model diagnostics (2 Warning) are listed in the Concerns chapter at the top of the analysis.

Data

A few columns, such as engine_location, display considerable differences in values with a measured difference of 21,644 between groups. Relationships such as the high association between compression_ratio and fuel_type might guide which variables to focus on when reviewing this dataset.

Most notable in the Data section:

  • Compression ratio and Fuel type go together (association 0.985)
  • City mpg and Highway mpg move together (correlation 0.972)
  • Curb weight and Length move together (correlation 0.881)

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 24 columns across 201 rows (20 numeric, 4 non-numeric): every column is at least 75% populated. Distributional statistics on numeric columns can be read at face value. 6 columns are strongly skewed: the mean of such a column is pulled toward its long tail, so the median is the better guide to a typical value.

Columns
NameRoleTypeNMissingMeanStdMinP25MedianP75MaxSkewnessCV
TargetTargetinteger201013,20779475118777510,29516,50045,4001.810.6017
AspirationInputbool2010
Body styleInputinteger20102.1940.84102234-0.48230.3833
BoreInputnumber1974 (1.99 %)3.3310.2712.5403.1503.3103.5903.940-0.032620.0813
City mpgInputinteger201025.186.4213192430490.68040.2551
Compression ratioInputnumber201010.1644.0057.0008.6009.0009.40023.0002.5840.394
Curb weightInputinteger20102556517148821692414292640660.70580.2024
Drive wheelsInputinteger20101.3330.55101122-0.033590.4131
Engine locationInputbool2010
Engine sizeInputinteger2010126.941.561981201413261.9790.3275

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

N columns
24
N high missing
0
N rows
201
N strongly skewed
6
Table branch
main
Table ID
a5f7653d-38e4-466c-8c9a-741c6e3ce985

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.

Aspiration

Aspiration: 2 distinct values, most often no.

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
Aspiration
StatisticValue
Typebool
Values present201
Distinct values2
Continuousno
Modefalse
Rows per value
VariableValueRowsShare
Aspirationfalse16582.09 %
Aspirationtrue3617.91 %

How to read this

The figures below summarize Aspiration 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 Aspiration 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 Aspiration
AspirationRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeight
false1655160.42.99498.9798.07172.91.3092.2063.2249.593.30212542.22.13965.592478.1124.725.752.33912.0931.4553.66
true364923.64.472123.4102.1180.31.4442.1393.40412.793.46116254.81.83367.262911.3136.622.582.2513.2527.1754.24

How to read this

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

Visualization

The charts in this chapter were chosen from the type of Aspiration 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.

Body style

Body style: 0 to 4, median 2, skewed to high values, 5 distinct values; moves opposite to Height most (r -0.482).

Description
Body style
StatisticValue
Typeinteger
Values present201
Distinct values5
Continuousyes
Decimal places0
Minimum0
Maximum4
Mean2.194
Median2
Mode2

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

By Value

Body style accounts for 14% of the spread in Target: 0 averages 12,251 above 3 (95% interval 70.9 to 24,431). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target0822208.510039.8 to 34377.2
Target4621890.510149.6 to 33631.4
Target29414459.812714 to 16205.5
Target1251237210258.1 to 14485.8
Target3689957.48953.2 to 10961.7
Mean per value of Body style
Body styleRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeight
085031.34.25142.398.5176.91.8753.32310.733.60922208.52.566.612810.6176.321.63314.6327.2552.85
1255014.63.1297.75102.2181.31.23.17610.323.40612372266.262784.2123.824.042.2414.7228.7256.73
29450753.436103.8100.81781.3513.27110.973.34514459.8266.322625.9131.725.052.45712.0230.5754.39
3685220.92.79497.6795.44166.31.253.289.0433.239957.42.20665.12322.9112.926.62.04411.6832.3852.13
465158.35131.792.7170.41.8333.0438.9333.49221890.51.83365.582801.7157.220.52.83310.52651.43
Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Heighttwo numbers0.5251-0.4819-0.5815 to -0.3681-0.5251-0.4248< 0.001201
Wheel basetwo numbers0.4484-0.3734-0.4867 to -0.2478-0.4484-0.356< 0.001201
Number of doorsa number and a category0.4471< 0.001199
Lengthtwo numbers0.4182-0.3886-0.5001 to -0.2644-0.4182-0.3222< 0.001201
Curb weighttwo numbers0.2927-0.2606-0.3851 to -0.1268-0.2927-0.2279< 0.001201
Boretwo numbers0.2428-0.2428-0.3701 to -0.1066-0.2426-0.19250.002245197
Widthtwo numbers0.2216-0.22-0.3479 to -0.08422-0.2216-0.17580.005574201
Targettwo numbers0.206-0.1795-0.3102 to -0.04215-0.206-0.16020.02755201
Engine sizetwo numbers0.1963-0.1866-0.3169 to -0.04952-0.1963-0.15660.02294201
Number of engine cylinderstwo numbers0.1787-0.1464-0.2792 to -0.008218-0.1787-0.1630.06729201

The first 10 rows of 23. The other 13 are in the notebook.

How to read this

Each row pairs Body style 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 Body style, 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 Body style'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 Body style (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 Body style 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.

Bore

Bore: 2.54 to 3.94, median 3.31, 38 distinct values, 2% missing; moves with Curb weight most (r 0.645).

Description
Bore
StatisticValue
Typenumber
Values present197
Distinct values38
Continuousyes
Decimal places2
Minimum2.54
Maximum3.94
Mean3.331
Median3.31
Mode3.62

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

By Value

Bore accounts for 45% of the spread in Target: 3.74 averages 28,126 above 2.97 (95% interval 22,617 to 33,634). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target3.74334,52828836.1 to 40219.9
Target3.58624,61517512.7 to 31717.3
Target3.54618716.79338 to 28095.3
Target3.46917963.910413.7 to 25514.1
Target3.78816734.114164.5 to 19303.7
Target3.7515,79713164.4 to 18429.6
Target3.43815760.813203.3 to 18318.2
Target3.31814313.48654.1 to 19972.7
Target3.59314040.711253.9 to 16827.5
Target3.622313796.39585.7 to 18006.9

The first 10 rows of 21. The other 11 are in the notebook.

Mean per value of Bore
BoreRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioTargetWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeightEngine type
060002.75109.595.3169239.413,02065.72411.372.516.7582349.65
2.5415250511099.1186.632.079.315,04066.52707121212172856.12
2.6815000515494.5171.2233.47916,50065.528231521942652.44
2.9175628.667.1491.31149.12.8573.3079.4296405.363.441823.385.7136.431.8574.71441.2952.441.857
2.92160007696.5157.13.419.2729563.920249230253458.32
2.971255006893.71592.6673.239.4016402.363.95196991.1732.529.08338.7550.732
2.991480035594.5165.323.4721.9709963.82017103452125054.52
3.0154680369.2100.3176.81.223.42312,31666.462528106.6342.420.239.855.542
3.031052802.88293.55158.82.73.2388.566981.163.922001.593.629.527.135.152.342
3.05648006295.7164.223.0396961.363.62286.79230.332193656.82

The first 10 rows of 39. The other 29 are in the notebook.

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Engine sizetwo numbers0.71970.58290.4824 to 0.66820.71970.5555< 0.001197
Curb weighttwo numbers0.70060.64460.5548 to 0.71960.70060.5352< 0.001197
Targettwo numbers0.64930.54340.4368 to 0.6350.64930.4928< 0.001197
Horsepowertwo numbers0.64630.56850.4652 to 0.65660.64630.4773< 0.001195
Lengthtwo numbers0.64230.61010.5141 to 0.69090.64230.5008< 0.001197
Highway mpgtwo numbers0.6321-0.5992-0.6819 to -0.5014-0.6321-0.4731< 0.001197
City mpgtwo numbers0.6193-0.5927-0.6764 to -0.4938-0.6193-0.4628< 0.001197
Widthtwo numbers0.60760.54490.4385 to 0.63630.60760.4465< 0.001197
Wheel basetwo numbers0.54140.49490.3815 to 0.59360.54140.4124< 0.001197
Drive wheelstwo numbers0.52410.47910.3637 to 0.58010.52410.4314< 0.001197

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

The rows of Bore are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of -0.08 between each value and the next), so the order carries no pattern to show.

City mpg

City mpg: 13 to 49, median 24, 29 distinct values; moves with Highway mpg most (r 0.972).

Description
City mpg
StatisticValue
Typeinteger
Values present201
Distinct values29
Continuousyes
Decimal places0
Minimum13
Maximum49
Mean25.18
Median24
Mode31

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

By Value

City mpg accounts for 72% of the spread in Target: 15 averages 28,134 above 38 (95% interval 22,868 to 33,401). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target15334893.328971.6 to 40815.1
Target16531,39219481.5 to 43302.5
Target22428,39424450.7 to 32337.3
Target171220738.414932.9 to 26543.9
Target18319028.313672.2 to 24384.5
Target20318937.76783.5 to 31091.8
Target192716248.915143.9 to 17354
Target21815,83012830.5 to 18829.5
Target231211997.49925.7 to 14069.1
Target28711557.77444.1 to 15671.3

The first 10 rows of 19. The other 9 are in the notebook.

Mean per value of City mpg
City mpgRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeNumber of engine cylindersMakeHighway mpgHeight
13150005262102191.7222.7611.53.5436,000470.63950326671747.8
14245005184116.5203.7213.3583.843,180471.853807.5306591656.05
1534966.75178112198.7223.918.0673.62734893.30.666770.033879241.745.33319.3353.97
16552605161.8102.9186.922.63.2458.43.5431,3923.468.543307193.23.6620.653.26
17125583.34.25157.796.74176.81.7523.1348.8173.54220738.43.566.572856.3143.82.58312.3322.9252.84
1835533.35130.3105.51841.33323.167.9333.4619028.32.33367.872988.7147.7311.672355.27
19275192.65.111134.1102182.11.5932.2963.2858.4153.46816248.91.81567.582938.7144.72.59311.9324.7453.77
2034883.35146103.6186.822.3333.2979.1673.28318937.70.666767.033054168.7413.3324.3353.93
2184937.5511397179.71.52.752.8789.1513.3415,8301.565.482673.61342.5927.7553.83
22443503123110.619321.253.6421.53.5828,394270.653632.5183392556.6

The first 10 rows of 29. The other 19 are in the notebook.

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Highway mpgtwo numbers0.9720.9720.9632 to 0.97880.96860.8942< 0.001201
Horsepowertwo numbers0.9119-0.8226-0.863 to -0.7718-0.9119-0.7691< 0.001199
Targettwo numbers0.8313-0.6866-0.7534 to -0.6057-0.8313-0.6613< 0.001201
Curb weighttwo numbers0.8056-0.7495-0.8045 to -0.6819-0.8056-0.6446< 0.001201
Engine sizetwo numbers0.7221-0.6505-0.7238 to -0.5628-0.7221-0.5843< 0.001201
Fuel systemtwo numbers0.7102-0.6637-0.7346 to -0.5784-0.7102-0.5414< 0.001201
Widthtwo numbers0.6734-0.6335-0.7097 to -0.5427-0.6734-0.5078< 0.001201
Lengthtwo numbers0.6652-0.6652-0.7358 to -0.5802-0.661-0.5003< 0.001201
Boretwo numbers0.6193-0.5927-0.6764 to -0.4938-0.6193-0.4628< 0.001197
Drive wheelstwo numbers0.5323-0.4692-0.5706 to -0.3538-0.5323-0.4358< 0.001201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

The rows of City mpg are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.15 between each value and the next), so the order carries no pattern to show.

Compression ratio

Compression ratio: 7 to 23, median 9, skewed to high values, 32 distinct values; moves opposite to Peak rpm most (r -0.436).

Description
Compression ratio
StatisticValue
Typenumber
Values present201
Distinct values32
Continuousyes
Decimal places2
Minimum7
Maximum23
Mean10.16
Median9
Mode9.0

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

By Value

Compression ratio accounts for 54% of the spread in Target: 8.3 averages 24,589 above 9.6 (95% interval 9,151 to 40,026). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target8.3331038.315589.7 to 46487
Target8830943.520935.5 to 40951.5
Target21.5428,39424450.7 to 32337.3
Target9.51318767.612473 to 25062.2
Target8.8318,28011357.6 to 25202.4
Target21515,79713164.4 to 18429.6
Target8.4514,64911776.1 to 17521.9
Target7614310.812103.6 to 16518
Target7.5513177.46920.8 to 19434
Target9.31113,07010935.3 to 15204.7

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

Mean per value of Compression ratio
Compression ratioRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeBoreTargetWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeightEngine type
7.0651005.5144.597.92175.51.3332.8333.7583.59314310.866.632906.5152.318.83210.672451.171.833
7.5553405.6134.499.5179.11.423.3363.3513177.466.122675.611820.621526.853.262
7.6455005.510293.53157.32.753.393.038040.363.82148982428.253050.752
7.724800511196.95172.81.52.643.6211476.565.4258010823.52182654.63
7.815200520091.3170.7233.273.4319,69967.93139181174122349.74
8.085087.55165.5107.1189.61.751.8753.3193.64930943.568.743266.8206.916.383.6255.37521.1355.162.5
8.1247505176113199.6224.173.6333,90069.6406625815471952.8
8.3350005150106191.91.6672.6673.23.3531038.371.23503.7199.716.334.3336.33318.6754.43.333
8.455000596.2110.4191.621.62.793.4614,64968.43137120192132457.1
8.5145228.62.71495.1498.93175.21.0712.1433.4193.27510,96865.842469.4123.123.212.28610.2929.2954.412

The first 10 rows of 32. The other 22 are in the notebook.

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Fuel typea number and a category0.9852< 0.001201
City mpgtwo numbers0.47570.33140.2023 to 0.44920.47570.3737< 0.001201
Peak rpmtwo numbers0.436-0.436-0.5422 to -0.3161-0.03087-0.006182< 0.001199
Highway mpgtwo numbers0.43360.26850.1351 to 0.39230.43360.3278< 0.001201
Horsepowertwo numbers0.356-0.2146-0.3434 to -0.07781-0.356-0.2760.006725199
Fuel systemtwo numbers0.3388-0.09993-0.2351 to 0.039-0.3388-0.25930.202201
Aspirationa number and a category0.3075< 0.001201
Heighttwo numbers0.25970.25970.1259 to 0.38430.008056-0.01269< 0.001201
Wheel basetwo numbers0.25030.25030.1159 to 0.3757-0.1195-0.10240.001112201
Engine sizetwo numbers0.23320.02889-0.1099 to 0.1666-0.2332-0.18430.7865201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

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

Curb weight

Curb weight: 1,488 to 4,066, median 2,414, skewed to high values; moves with Length most (r 0.881).

Description
Curb weight
StatisticValue
Typeinteger
Values present201
Distinct values169
Continuousyes
Decimal places0
Minimum1488
Maximum4066
Mean2555.7
Median2414
Mode2385

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
Targettwo numbers0.91380.83440.7869 to 0.87210.91380.7443< 0.001201
Lengthtwo numbers0.88980.88070.8453 to 0.90840.88980.7222< 0.001201
Engine sizetwo numbers0.87440.84910.8053 to 0.88360.87440.7209< 0.001201
Widthtwo numbers0.86620.86620.8269 to 0.89710.85860.6802< 0.001201
Highway mpgtwo numbers0.8309-0.7949-0.8408 to -0.7376-0.8309-0.6706< 0.001201
City mpgtwo numbers0.8056-0.7495-0.8045 to -0.6819-0.8056-0.6446< 0.001201
Horsepowertwo numbers0.80330.75810.6919 to 0.81160.80330.6166< 0.001199
Wheel basetwo numbers0.78210.78210.7218 to 0.83060.76460.6096< 0.001201
Boretwo numbers0.70060.64460.5548 to 0.71960.70060.5352< 0.001197
Fuel systemtwo numbers0.64450.60260.5065 to 0.6840.64450.484< 0.001201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

The rows of Curb weight are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.06 between each value and the next), so the order carries no pattern to show.

Drive wheels

Drive wheels: 0 to 2, median 1, skewed to high values, 3 distinct values; moves with Curb weight most (r 0.591).

Description
Drive wheels
StatisticValue
Typeinteger
Values present201
Distinct values3
Continuousyes
Decimal places0
Minimum0
Maximum2
Mean1.333
Median1
Mode1

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

By Value

Drive wheels accounts for 40% of the spread in Target: 2 averages 10,513 above 1 (95% interval 8,340 to 12,685). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target27519757.617667.9 to 21847.3
Target0810,2417492 to 12990
Target11189244.88634.8 to 9854.8
Mean per value of Drive wheels
Drive wheelsRowsPeak rpmFuel systemHorsepowerWheel baseLengthBody styleStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeight
084837.52.587.2596.49170.61.6252.8338.5133.42410,2412.62564.882553.6107.5242.12516.1327.8855.83
11185172.42.54286.2596.461692.3143.2919.9393.29244.82.06865.082270.7109.128.152.08511.9834.0853.54
2755062.74.467131.6102.7182.82.0673.24810.693.53719757.62.05367.273004.2156.920.632.7212.3925.6553.9
Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Targettwo numbers0.64390.58830.4898 to 0.6720.64390.5134< 0.001201
Curb weighttwo numbers0.61180.59060.4925 to 0.67390.61180.4857< 0.001201
Horsepowertwo numbers0.60230.5470.4416 to 0.63760.60230.4899< 0.001199
Engine sizetwo numbers0.56590.5260.4181 to 0.61930.56590.4636< 0.001201
Highway mpgtwo numbers0.5426-0.4729-0.5737 to -0.3579-0.5426-0.4439< 0.001201
City mpgtwo numbers0.5323-0.4692-0.5706 to -0.3538-0.5323-0.4358< 0.001201
Boretwo numbers0.52410.47910.3637 to 0.58010.52410.4314< 0.001197
Widthtwo numbers0.51450.48140.3675 to 0.58110.51450.4248< 0.001201
Lengthtwo numbers0.49720.49520.3831 to 0.5930.49720.4027< 0.001201
Wheel basetwo numbers0.46770.46770.3521 to 0.56920.41120.3353< 0.001201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

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

Engine location

Engine location: 2 distinct values, most often no.

Description
Engine location
StatisticValue
Typebool
Values present201
Distinct values2
Continuousno
Modefalse
Rows per value
VariableValueRowsShare
Engine locationfalse19898.51 %
Engine locationtrue31.493 %
By Value

Engine location accounts for 10% of the spread in Target: yes averages 21,644 above no (95% interval 16,815 to 26,473). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Targettrue334,52828836.1 to 40219.9
Targetfalse19812884.111825.6 to 13942.5
Mean per value of Engine location
Engine locationRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeight
false1985105.63.232101.898.94174.31.3232.2073.26210.173.32412884.12.07165.92552.4125.925.32.29812.2630.7753.8
true35900520789.5168.921.3332.99.53.7434,5283652770.7194174152551.6

Engine size

Engine size: 61 to 326, median 120, 43 distinct values; moves with Target most (r 0.872).

Description
Engine size
StatisticValue
Typeinteger
Values present201
Distinct values43
Continuousyes
Decimal places0
Minimum61
Maximum326
Mean126.9
Median120
Mode92

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

By Value

Engine size accounts for 89% of the spread in Target: 209 averages 29,934 above 90 (95% interval 16,836 to 43,032). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target209336318.323152.7 to 49483.9
Target194334,52828836.1 to 40219.9
Target183428,39424450.7 to 32337.3
Target164322213.317151.4 to 27275.3
Target130416841.312916.3 to 20766.2
Target141716767.113698.2 to 19836.1
Target136516,52513744.8 to 19305.2
Target181616115.713304.9 to 18926.4
Target171316,08214988.8 to 17175.2
Target152615914.213901.4 to 17927

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

Mean per value of Engine size
Engine sizeRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetWidthCurb weightCity mpgNumber of engine cylindersMakeHighway mpgHeightEngine type
61151004888.4141.133.039.52.91515160.314884735353.2
7036000210195.3169239.412,14565.72381.71782349.65
79155006093.715033.0710.12.91539964183738254252.62
8016000513595.3169239.415,64565.725001682349.65
9010548068.493.86158.32.63.2069.4412.9826384.263.761935.433.626.839.650.962
91550006893.1162.22.63.1593.04643564.2191830.82836.654.12
9215534067.693.93157.82.43.2229.1872.9796773.163.92069.932.07211.837.454.192
97145021.41.57166.4395.51167.71.8573.25513.263.1447815.664.362088.432.43214.7138.6454.432.071
98145307.12.8578594.59164.51.2862.53.1548.6863.1368170.964.072161.927.93215.5733.7952.041.714
1031480035594.5165.323.4721.92.99709963.82017452125054.52

The first 10 rows of 43. The other 33 are in the notebook.

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Curb weighttwo numbers0.87440.84910.8053 to 0.88360.87440.7209< 0.001201
Targettwo numbers0.87230.87230.8347 to 0.90190.82840.6613< 0.001201
Number of engine cylinderstwo numbers0.83570.83570.7885 to 0.87310.68820.5811< 0.001201
Horsepowertwo numbers0.82270.82270.772 to 0.8630.81730.6588< 0.001199
Lengthtwo numbers0.77970.6850.6039 to 0.75210.77970.5993< 0.001201
Widthtwo numbers0.7630.72940.6574 to 0.78830.7630.5848< 0.001201
City mpgtwo numbers0.7221-0.6505-0.7238 to -0.5628-0.7221-0.5843< 0.001201
Boretwo numbers0.71970.58290.4824 to 0.66820.71970.5555< 0.001197
Highway mpgtwo numbers0.7168-0.6796-0.7477 to -0.5974-0.7168-0.5797< 0.001201
Wheel basetwo numbers0.64570.5720.4709 to 0.65830.64570.5013< 0.001201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

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

Engine type

Engine type: 0 to 5, median 2, 6 distinct values.

Description
Engine type
StatisticValue
Typeinteger
Values present201
Distinct values6
Continuousyes
Decimal places0
Minimum0
Maximum5
Mean2.085
Median2
Mode2

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

By Value

Engine type accounts for 18% of the spread in Target: 4 averages 13,531 above 2 (95% interval 6,561 to 20,501). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target41325098.418183.5 to 32013.3
Target01218116.413056.4 to 23176.5
Target11214627.612293.5 to 16961.7
Target31513738.67685.5 to 19791.7
Target5413,0209711.7 to 16328.3
Target214511567.410509.4 to 12625.3
Mean per value of Engine type
Engine typeRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeight
01253755146100.4182.51.8332.5833.28393.40318116.466.42976.8157.320313.524.8352.73
1124704.23.83395.5108.41871.9171.753.14913.633.52714627.667.723076.6129.624.51.91712.1728.8356.85
21455127.62.96693.5798.03172.11.2342.2143.32910.243.26511567.465.542425.3117.826.722.16611.9732.553.82
31550002.867110.494.84168.90.86671.82.6738.9533.64413738.664.962407.1124.524.472.417.429.653.32
4135015.45169.5102.6185.61.7692.2313.2028.8233.45425098.468.783355.8216.716.774.46210.4621.2353.12
5460002.75109.595.3169239.413,02065.72411.372.516.7582349.6
Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Stroketwo numbers0.2244-0.1894-0.3207 to -0.05094-0.2244-0.17780.08843197
Heighttwo numbers0.2058-0.1951-0.3247 to -0.05825-0.2058-0.16560.08843201
Wheel basetwo numbers0.2013-0.1439-0.2768 to -0.005643-0.2013-0.16320.1602201
City mpgtwo numbers0.1715-0.1715-0.3027 to -0.03393-0.1371-0.10480.1143201
Highway mpgtwo numbers0.1565-0.1565-0.2886 to -0.01846-0.101-0.078220.1527201
Number of engine cylinderstwo numbers0.15110.111-0.02786 to 0.24560.15110.14410.2238201
Drive wheelstwo numbers0.1503-0.06734-0.2038 to 0.07173-0.1503-0.14830.4558201
Horsepowertwo numbers0.14450.14450.005477 to 0.2780.076620.054650.1602199
Lengthtwo numbers0.1413-0.08967-0.2253 to 0.04934-0.1413-0.1130.306201
Aspirationa number and a category0.13090.1825201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

The rows of Engine type are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.10 between each value and the next), so the order carries no pattern to show.

Fuel system

Fuel system: 0 to 7, median 5, skewed to low values, 8 distinct values; moves with Horsepower most (r 0.675).

Description
Fuel system
StatisticValue
Typeinteger
Values present201
Distinct values8
Continuousyes
Decimal places0
Minimum0
Maximum7
Mean3.259
Median5
Mode5

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

By Value

Fuel system accounts for 33% of the spread in Target: 5 averages 10,172 above 1 (95% interval 8,347 to 11,997). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target59217605.115811.2 to 19399
Target32015838.212206.4 to 19469.9
Target2312,1458729.9 to 15560.1
Target6910990.48883 to 13097.9
Target0117555.56621.6 to 8489.4
Target1647433.27087.8 to 7778.6
Mean per value of Fuel system
Fuel systemRowsPeak rpmHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeight
0115772.776.5593.94158.72.5453.4419.2452.9987555.5264.162045.597.3631.452536.5553.36
1645048.473.1995.33165.60.96882.2663.159.0723.1967433.22.10964.312127.7100.529.911.98412.2335.853.49
23600010195.3169239.412,145565.72381.7701782349.6
320443584.45104.41821.551.83.48622.013.37515838.21.7567.432898.8135.730.32.314.3534.7555.84
41500014595.9173.233.973.612,964266.3281115619242450.2
5925195.6129.9101.11811.5762.0873.2178.813.45217605.12.06566.912841.4148.320.742.71713.1326.2254.07
695277.8125.895.47169.51.1112.7783.6227.33.32410990.4265.442549.7127.821.44211.6727.3350.56
7150009096172.6233.239.23.4311,048265.2273411924262951.4
Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Horsepowertwo numbers0.75540.67540.5919 to 0.74460.75540.5678< 0.001199
City mpgtwo numbers0.7102-0.6637-0.7346 to -0.5784-0.7102-0.5414< 0.001201
Highway mpgtwo numbers0.6822-0.638-0.7134 to -0.548-0.6822-0.5091< 0.001201
Targettwo numbers0.67930.52170.4131 to 0.61560.67930.5067< 0.001201
Curb weighttwo numbers0.64450.60260.5065 to 0.6840.64450.484< 0.001201
Engine sizetwo numbers0.60420.50730.3967 to 0.60330.60420.4569< 0.001201
Widthtwo numbers0.55340.51340.4037 to 0.60850.55340.4165< 0.001201
Lengthtwo numbers0.55110.55110.4468 to 0.64070.53940.4138< 0.001201
Boretwo numbers0.48170.47320.3571 to 0.5750.48170.3762< 0.001197
Drive wheelstwo numbers0.43590.43590.3166 to 0.54160.42580.387< 0.001201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

The rows of Fuel system 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.

Fuel type

Fuel type: 2 distinct values, most often yes.

Description
Fuel type
StatisticValue
Typebool
Values present201
Distinct values2
Continuousno
Modetrue
Rows per value
VariableValueRowsShare
Fuel typetrue18190.05 %
Fuel typefalse209.95 %
By Value
Mean per value of Fuel type
Fuel typeRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpgHeight
false204435384.45104.41821.551.83.48622.013.37515838.21.7567.432898.8135.730.32.314.3534.7555.84
true1815193.93.287105.598.18173.31.3092.2383.2318.8563.32612916.42.12265.722517.8125.924.612.32612.0730.2453.54

Height

Height: 47.8 to 59.8, median 54.1, 49 distinct values; moves with Wheel base most (r 0.591).

Description
Height
StatisticValue
Typenumber
Values present201
Distinct values49
Continuousyes
Decimal places1
Minimum47.8
Maximum59.8
Mean53.77
Median54.1
Mode50.8

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

By Value

Height accounts for 17% of the spread in Target: 51.6 averages 12,981 above 54.5 (95% interval 3.29 to 25,958). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target51.6719617.16640 to 32594.3
Target56.7818726.911026.7 to 26427
Target49.7318432.315326.3 to 21538.3
Target58.7417905.86509 to 29302.5
Target52.8616733.72718.3 to 30749.1
Target57.5316293.39402 to 23184.7
Target55.5915916.711493.4 to 20339.9
Target54.3815915.312363.8 to 19466.7
Target56.2315781.78961.1 to 22602.2
Target56.1715105.612673.2 to 17538

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

Mean per value of Height
HeightRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHighway mpg
47.8150005262102191.7222.7611.53.5436,000470.63950326136717
48.825000511188.6168.8242.6893.4714997.564.1254813021227
49.4252503.510296.317333.4683.269229265.423491162421131
49.6460002.75109.595.3169239.413,020565.72411.372.516.75823
49.7352005173.393.93173.3233.278.63.4318432.3467.93116.318118.3341224.33
50.265083.35.5144.795.67172.51.33333.7427.4173.64814955.5266.632847.8155.21921124.5
50.51596.1176.833.98.73.469895266.624601322321631
50.6555001.874.893.7157.32.43.2629.042.9827063.4263.82020.691.629.62636.4
50.8145432.12.14381.3692.84158.61.0712.9293.2818.9363.0098718.72.14364.422095.3103.3312.2149.35736.64
51.01550010096.5169.123.589.13.1510,3452662293110252531

The first 10 rows of 49. The other 39 are in the notebook.

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Wheel basetwo numbers0.64140.59070.4926 to 0.6740.64140.4681< 0.001201
Number of doorsa number and a category0.546< 0.001199
Lengthtwo numbers0.53220.49210.3795 to 0.59030.53220.381< 0.001201
Body styletwo numbers0.5251-0.4819-0.5815 to -0.3681-0.5251-0.4248< 0.001201
Widthtwo numbers0.37060.3060.175 to 0.42630.37060.272< 0.001201
Curb weighttwo numbers0.36290.30760.1767 to 0.42780.36290.2544< 0.001201
Peak rpmtwo numbers0.3116-0.3116-0.4319 to -0.1803-0.2838-0.2002< 0.001199
Fuel typea number and a category0.2816< 0.001201
Maketwo numbers0.26420.23080.0954 to 0.35770.26420.19350.002258201
Targettwo numbers0.26380.1355-0.002967 to 0.26880.26380.18240.09757201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

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

Highway mpg

Highway mpg: 16 to 54, median 30, 30 distinct values; moves with City mpg most (r 0.972).

Description
Highway mpg
StatisticValue
Typeinteger
Values present201
Distinct values30
Continuousyes
Decimal places0
Minimum16
Maximum54
Mean30.69
Median30
Mode25

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

By Value

Highway mpg accounts for 62% of the spread in Target: 25 averages 16,833 above 41 (95% interval 13,043 to 20,623). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target251922344.218555.9 to 26132.6
Target22722113.312692.4 to 31534.2
Target27518552.613856.3 to 23248.9
Target26317756.714990.9 to 20522.5
Target281215440.813464.6 to 17417.1
Target23714,99411148.9 to 18839.1
Target241714876.713911.7 to 15841.7
Target33911405.98486.8 to 14325
Target291011087.18730.1 to 13444.1
Target301610060.38685.3 to 11435.2

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

Mean per value of Highway mpg
Highway mpgRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetWidthCurb weightEngine sizeCity mpgNumber of engine cylindersMakeHeightEngine type
16245005184116.5203.7213.3583.843,18071.853807.5306145956.054
17150005262102191.7222.7611.53.5436,00070.63950326136747.84
18247505155106.1191.5233.18.33.4634,62071.13712.5234165953.654
19247505176113199.6224.178.13.6333,90069.64066258154752.8
20254505161107.9194.91.523.3958.153.37530377.571.153295.5170163.51.556.12
2275271.45158.1102.3186.21.4291.7143.2898.1433.50422113.366.943146.716816.863.28610.1455.512.571
2375642.93.714126.196.93172.71.7142.5712.9278.8433.54314,99466.432635.1107.417.861.42911.8651.214.429
24175111.85.176133.4103.41831.7062.2943.3198.0883.46514876.767.362998.2144.919.062.52913.8253.81.176
25195013.24.158135.9103.2184.81.6841.6843.28612.913.54222344.268.23099.1164.420.263.10510.9554.662.421
2635333.3515897.57181.51.3332.6673.20393.25317756.766.172826131.3192.66711.3354.871.333

The first 10 rows of 30. The other 20 are in the notebook.

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
City mpgtwo numbers0.9720.9720.9632 to 0.97880.96860.8942< 0.001201
Horsepowertwo numbers0.8866-0.8046-0.8487 to -0.7494-0.8866-0.7344< 0.001199
Curb weighttwo numbers0.8309-0.7949-0.8408 to -0.7376-0.8309-0.6706< 0.001201
Targettwo numbers0.8273-0.7047-0.7682 to -0.6275-0.8273-0.6518< 0.001201
Engine sizetwo numbers0.7168-0.6796-0.7477 to -0.5974-0.7168-0.5797< 0.001201
Lengthtwo numbers0.6981-0.6981-0.7628 to -0.6196-0.6888-0.5241< 0.001201
Widthtwo numbers0.6925-0.6806-0.7485 to -0.5986-0.6925-0.5185< 0.001201
Fuel systemtwo numbers0.6822-0.638-0.7134 to -0.548-0.6822-0.5091< 0.001201
Boretwo numbers0.6321-0.5992-0.6819 to -0.5014-0.6321-0.4731< 0.001197
Wheel basetwo numbers0.5433-0.5433-0.634 to -0.4378-0.5311-0.4035< 0.001201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

The rows of Highway mpg are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.14 between each value and the next), so the order carries no pattern to show.

Horsepower

Horsepower: 48 to 262, median 95, skewed to high values, 1% missing; moves with Engine size most (r 0.823).

Description
Horsepower
StatisticValue
Typeinteger
Values present199
Distinct values58
Continuousyes
Decimal places0
Minimum48
Maximum262
Mean103.4
Median95
Mode68

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
City mpgtwo numbers0.9119-0.8226-0.863 to -0.7718-0.9119-0.7691< 0.001199
Highway mpgtwo numbers0.8866-0.8046-0.8487 to -0.7494-0.8866-0.7344< 0.001199
Targettwo numbers0.85050.81050.7568 to 0.85340.85050.6665< 0.001199
Engine sizetwo numbers0.82270.82270.772 to 0.8630.81730.6588< 0.001199
Curb weighttwo numbers0.80330.75810.6919 to 0.81160.80330.6166< 0.001199
Fuel systemtwo numbers0.75540.67540.5919 to 0.74460.75540.5678< 0.001199
Widthtwo numbers0.68190.61530.5208 to 0.69490.68190.5064< 0.001199
Number of engine cylinderstwo numbers0.67090.67090.5865 to 0.74090.56540.4647< 0.001199
Lengthtwo numbers0.65980.58030.48 to 0.66570.65980.4782< 0.001199
Boretwo numbers0.64630.56850.4652 to 0.65660.64630.4773< 0.001195

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

The rows of Horsepower are not charted in their order: consecutive rows are no more alike than rows picked at random (a correlation of 0.10 between each value and the next), so the order carries no pattern to show.

Length

Length: 141 to 208, median 173; moves with Curb weight most (r 0.881).

Description
Length
StatisticValue
Typenumber
Values present201
Distinct values73
Continuousyes
Decimal places1
Minimum141.1
Maximum208.1
Mean174.2
Median173.2
Mode157.3

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
Wheel basetwo numbers0.91310.8760.8394 to 0.90470.91310.747< 0.001201
Widthtwo numbers0.88980.85720.8155 to 0.890.88980.7257< 0.001201
Curb weighttwo numbers0.88980.88070.8453 to 0.90840.88980.7222< 0.001201
Targettwo numbers0.81030.69060.6106 to 0.75670.81030.6364< 0.001201
Engine sizetwo numbers0.77970.6850.6039 to 0.75210.77970.5993< 0.001201
Highway mpgtwo numbers0.6981-0.6981-0.7628 to -0.6196-0.6888-0.5241< 0.001201
City mpgtwo numbers0.6652-0.6652-0.7358 to -0.5802-0.661-0.5003< 0.001201
Horsepowertwo numbers0.65980.58030.48 to 0.66570.65980.4782< 0.001199
Boretwo numbers0.64230.61010.5141 to 0.69090.64230.5008< 0.001197
Fuel systemtwo numbers0.55110.55110.4468 to 0.64070.53940.4138< 0.001201

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

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

Make

Make: 0 to 21, median 12, 22 distinct values.

Description
Make
StatisticValue
Typeinteger
Values present201
Distinct values22
Continuousyes
Decimal places0
Minimum0
Maximum21
Mean12.3
Median12
Mode19

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

By Value

Make accounts for 77% of the spread in Target: 7 averages 28,593 above 3 (95% interval 24,131 to 33,055). An association, not a cause.

Groups
TargetValueRowsAverage or share95% interval
Target7334,60029513.6 to 39686.4
Target9833,64727970.8 to 39323.2
Target15431400.521011.7 to 41789.3
Target2826118.818374 to 33863.5
Target211118063.215836.4 to 20290
Target1617859.214236.1 to 21482.2
Target0315498.311188.5 to 19808.2
Target131115489.113979.7 to 16998.5
Target17615223.312221.1 to 18225.6
Target81710652.98608.8 to 12697

The first 10 rows of 19. The other 9 are in the notebook.

Mean per value of Make
MakeRowsPeak rpmFuel systemHorsepowerWheel baseLengthDrive wheelsBody styleStrokeCompression ratioBoreTargetEngine typeWidthCurb weightEngine sizeCity mpgNumber of engine cylindersHighway mpgHeight
0350005125.390.57169.623.6672.94393.20715498.31.33364.572639.7137.320.332.66726.6750
1655005114.5102.7184.81.8333.48.6333.1817859.2268.852758.7130.719.332.83324.554.83
285068.85138.9103.2184.5223.1688.5753.47426118.8266.482929.4166.919.383.525.3854.83
33530062.6792.47151.92.6673.0839.5672.9960071.66762.5175780.33411.66746.3352.4
495388.92.22286.3395.011612.4443.3668.6343.0947875.4264.172151.3102.728234.1151.64
5135753.80.461580.2394.33160.82.4623.4629.2153.0228184.7264.382096.899.3130.38235.4653.24
62490048495.15171.722.53.238.853.378916.5263.52535.51152422952.45
734833.35204.7109.3197223.79.2333.634,6001.33369.934027.3280.714.334.66718.3351.13
8175108.81.88285.5397.02170.81.3532.5883.29910.493.28710652.92.70665.592297.810325.711.52931.9453.36
984487.54146.3110.9195.321.6253.43314.833.60533,647371.063696.3226.518.542155.73

The first 10 rows of 22. The other 12 are in the notebook.

Correlation
Correlations
ColumnPair ofAssociationPearson's r95% CI (Pearson)Spearman's ρKendall's τq-valueRows
Heighttwo numbers0.26420.23080.0954 to 0.35770.26420.19350.01129201
Boretwo numbers0.23790.23790.1015 to 0.36560.23680.16720.01129197
Peak rpmtwo numbers0.2206-0.2134-0.3423 to -0.07657-0.2206-0.14670.01899199
Stroketwo numbers0.2047-0.2047-0.335 to -0.06686-0.1961-0.13280.02243197
Compression ratiotwo numbers0.16930.1326-0.005953 to 0.26610.16930.11370.1396201
Fuel systemtwo numbers0.1670.14480.006502 to 0.27760.1670.13750.1159201
Targettwo numbers0.1636-0.1636-0.2954 to -0.02583-0.01776-0.0009790.09325201
Wheel basetwo numbers0.16360.07393-0.06513 to 0.21020.16360.12060.4827201
Body styletwo numbers0.1599-0.1537-0.286 to -0.01564-0.1599-0.12360.09651201
Number of doorsa number and a category0.15640.09651199

The first 10 rows of 23. The other 13 are in the notebook.

Visualization

The rows of Make 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.

Number of engine cylinders

Number of cylinders: 0 to 6, median 2, skewed to high values, 7 distinct values; moves with Engine size most (r 0.836).

Description
Number of engine cylinders
StatisticValue
Typeinteger
Values present201
Distinct values7
Continuousyes
Decimal places0
Minimum0
Maximum6
Mean2.323
Median2
Mode2

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

By Value

Number of cylinders accounts for 60% of the spread in Target: 5 averages 28,597 above 2 (95% interval 20,274 to 36,920). An association, not a cause.

This section continues in the notebook.

Implications

Quality differences by drive_wheels suggest model results for this variable's third quarter deserve extra attention. The overall share of variation explained is 0.851 and deployment readiness is pending. One concern was flagged, see the Concerns chapter.

Most notable in the Implications section:

  • The model is far less accurate on rows with Drive wheels in the third quarter (1) (48 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 model for Target in this analysis uses only Engine size, and its output rises as Engine size increases. While the model accounts for a large share of the variation in Target and tracks Engine size closely, it falls just short of matching the predictive strength of a straight line through the data, mirroring its performance in most cases but not offering a consistent improvement. The model's answer for a typical move in Engine size, such as going from 98.0 to 141, is to raise the model's output by about 5,962 on average, with everything else as observed.

As the model does not outperform a straight line and was closer than that benchmark on less than half the held-out rows, its results should be seen as tracking the main trend in the data rather than sharpening predictions for individual cases. Accuracy also drops noticeably for higher values of Target and for the rows where Drive wheels is in the third quarter, indicating less reliable predictions in those segments. Before acting on this model, prioritize checking its predictions on new or high-magnitude cases and review performance in segments with lower accuracy.

Recommended Next Data

  • Collect more rows where Drive wheels is in the third quarter (1). Reason: the model is far less accurate on the 23 held-out rows where Drive wheels is in the third quarter (1), explaining 37% of the variation there.
  • Add columns that record details about aspiration, fuel type, and number of doors. Reason: fairness diagnostics found a prediction parity gap for aspiration and that the model's errors differ between groups for fuel type and number of doors.
  • Collect more rows with higher-magnitude values of the model's output. Reason: model accuracy degrades on higher-magnitude predictions, as residual variance grows with predicted value.
  • Add columns that record other characteristics of the data. Reason: the model does not beat a straight line through the data on Target, indicating that further predictors could increase model quality.

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.
Target by Input Range

Segment drive_wheels in the third quarter (1) is much weaker than the aggregate (23 held-out rows): segment R² 0.372 against aggregate R² 0.851 (gap -0.479). 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
Drive wheels = q3230.3719-0.479Much weaker than on all rows
Compression ratio = q4200.7530-0.098In line with all rows
Fuel system = q2220.7914-0.059In line with all rows
Engine type = q4360.8261-0.025In line with all rows
Number of doors = true210.8327-0.018In line with all rows
Engine location = false400.8508+0.000In line with all rows
Fuel type = true340.8621+0.011In line with all rows
Number of engine cylinders = q4390.8743+0.024In line with all rows
Aspiration = false340.8752+0.024In line with all rows
Highway mpg = q38-6.7179-7.569Too few rows to judge

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

Aggregate R²
0.8508
Measure
r2
Target
Target

Rows unlike the ones the model learned from

2 of 40 held-out rows (5%) lie further from the training rows than 99 in 100 rows like them would, measured on the 1 input the model reads (engine_size). 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
TargetEngine size2.5764025 %
Score bins · 24 of 30 rows shown
CountFromTo
500.149
10.1490.297
60.2970.446
20.4460.595
90.5950.744
80.7440.892
20.8921.041
11.0411.19
01.191.339
31.3391.487

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.

Target Fairness

Column Aspiration splits the held-out rows into 2 groups. The prediction-parity ratio is 0.685 (smallest over largest mean prediction), the largest absolute group residual mean is 647 (0.179 times the overall residual standard deviation). Parity gap: at least one group receives systematically smaller predictions on average. Whether that is a property of the target itself or a model bias against the group needs checking.

Aggregate residual std
3614.3
N columns
3
N proxies
2
Proxy threshold
0.5
Columns
Bias max abs residual meanColumnGroupsInterpretationN groupsN proxiesPrediction parity ratioVerdict
647.3Aspirationtrue, falseColumn Aspiration splits the held-out rows into 2 groups. The prediction-parity ratio is 0.685 (smallest over largest mean prediction), the largest absolute group residual mean is 647 (0.179 times the overall residual standard deviation). Parity gap: at least one group receives systematically smaller predictions on average. Whether that is a property of the target itself or a model bias against the group needs checking.20.6854Prediction parity gap
1704.2Fuel typefalse, trueColumn Fuel type splits the held-out rows into 2 groups. The prediction-parity ratio is 0.962 (smallest over largest mean prediction), the largest absolute group residual mean is 1,704 (0.472 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. compression_ratio goes closely with fuel_type (association 0.985 on a scale from 0 to 1): a model that reads compression_ratio can tell the groups of fuel_type apart, so leaving fuel_type out of the inputs does not keep it out of the predictions.210.9618Residual bias
1130.1Number of doorsfalse, trueColumn Number of doors splits the held-out rows into 2 groups. The prediction-parity ratio is 0.931 (smallest over largest mean prediction), the largest absolute group residual mean is 1,130 (0.313 times the overall residual standard deviation). Residual bias: at least one group is systematically over- or under-predicted. The model's errors are not zero-mean across groups; its predictions for the affected groups deserve caution. height goes closely with number_of_doors (association 0.546 on a scale from 0 to 1): a model that reads height can tell the groups of number_of_doors apart, so leaving number_of_doors out of the inputs does not keep it out of the predictions.210.9312Residual bias

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: target
  • Rows: 201 rows
  • Per-Segment Predictions: not produced, because no segment column was chosen
  • Fairness: compared across 4 categorical columns (aspiration, engine_location, fuel_type, number_of_doors); which of them are protected attributes is not decided here
  • Out-of-Distribution: held-out rows are scored against the training rows; predictions made outside this report are not scored
  • Provenance & Audit: recorded in the chapter of that name

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

NameStatusDetail
GeneralizationPassheld-out quality holds up against training quality
Prediction Interval CoverageWarnintervals are wider than their level requires, largest gap 10.00 percentage points
Baseline Lift for TargetWarnis not told apart from the best baseline, R² lift -0.0264
Model ConstraintsPassevery held-out prediction is finite and within the training range
FairnessWarn3 columns compared, 3 flagged: Aspiration (a group receives systematically smaller predictions), Fuel type (a group is systematically over- or under-predicted), Number of doors (a group is systematically over- or under-predicted)
Out-of-DistributionPass5% of held-out rows are out of distribution
N critical concerns
0
N fail
0
N pass
3
N warn
3
Overall verdict
Caution

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 IDed1557d0-48b7-11f1-8087-d378640fe787
Run time2026-10-25 01:16 UTC
Total rows201
Modeled targetsTarget
Segment columnnot available
Time limit1 h 0 min
Time taken1 h 19 min
Credit budget500 credits
Ufinq version0.3.45

Findings

1 finding from the implications diagnostics (1 Warning) is listed in the Concerns chapter at the top of the analysis.

The report continues in the notebook.

Loading Diafunc…