Heteroscedasticity
Heteroscedasticity is the condition in which response or regression-error variance changes across levels of predictors or fitted values.
Effective Data Analysis
Heteroscedasticity is the condition in which the spread of residuals varies across the range of predicted or explanatory values.
- How it is used
- Diagnoses regression fit by comparing residual spread with predicted values.
- How deep it goes
- Explains visual and histogram checks plus corrective responses, without developing formal heteroscedasticity tests.
- Examples
- Uses housing-price predictions, residual scatterplots, and residual histograms as diagnostic examples.
- What you can do after
- Diagnose/debug with it by plotting residuals, recognizing uneven accuracy, and revising the model.
Read how Effective Data Analysis explains heteroscedasticity →
Grokking Statistics
Heteroscedasticity is the condition in which the variance of a response changes across the values of an independent variable.
- How it is used
- Checks whether Pearson correlation or linear regression remains trustworthy when response variance changes across the independent variable.
- How deep it goes
- Defines the assumption, contrasts steady and funnel-shaped variance, and discusses transformations without developing an alternative model.
- Examples
- Uses funnel-shaped comparison plots, Anscombe’s quartet, and binned bird-strike altitude data with logarithmic transformation.
- What you can do after
- Evaluate correlation and regression claims by plotting variance patterns and testing whether logarithmic transformations improve the relationship.
Statistics Every Programmer Needs
Heteroscedasticity is the condition that the variance of regression residuals changes systematically across levels of the predictors.
- How it is used
- Tests whether linear-regression residual variance remains constant so model inferences remain reliable.
- How deep it goes
- Defines the assumption, explains inferential consequences, and gives visual and Breusch–Pagan diagnostics without discussing corrective methods.
- Examples
- Uses a fitted-values-versus-residuals plot and Python statsmodels calls for the Breusch–Pagan test.
- What you can do after
- Diagnose/debug with residual plots and the Breusch–Pagan test whether constant-variance assumptions hold.
Read how Statistics Every Programmer Needs explains heteroscedasticity →
Machine Learning System Design
Heteroscedasticity is the condition in which regression-error variance changes across predictor or fitted-value levels.
- How it is used
- Checks regression-model assumptions during error analysis before interpreting fitted results.
- How deep it goes
- Explains changing residual variance and identifies it through scale-location and related diagnostic plots.
- Examples
- Uses demand forecasting and a scale-location plot where higher fitted values have higher residual variance.
- What you can do after
- Diagnose/debug regression assumptions by reading residual, scale-location, Q-Q, and leverage plots.
Read how Machine Learning System Design explains heteroscedasticity →
Related concepts
- homoscedasticitycontrasts with · Homoscedasticity is the statistical condition in which regression-error or response variance remains approximately constant across predictor or fitted-value levels.
- Pearson correlation coefficientcontrasts with · The Pearson correlation coefficient measures the strength and direction of the linear relationship between two numerical variables.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.