EasyDeepLearn

How do you detect and fix heteroscedasticity?

hard

Answer

  • Heteroscedasticity = error variance depends on x.
  • Detect with a residuals-vs-fitted plot (funnel shape), Breusch-Pagan or White test.
  • Consequences: OLS coefficients remain unbiased but standard errors are wrong, so t-tests and CIs are invalid.
  • Fixes: (1) transform y (log, sqrt, Box-Cox) if variance increases with the mean; (2) use weighted least squares with weights proportional to 1/variance; (3) use heteroscedasticity-robust (sandwich / White) standard errors.
Check yourself — multiple choice
  • Heteroscedasticity means the errors are perfectly Gaussian
  • It's non-constant error variance; fix with a transform, WLS, or robust SEs
  • It biases the coefficient estimates
  • It only affects logistic regression

Non-constant variance → wrong SEs. Fix with transforms, WLS, or robust SEs.

#linear-regression#assumptions

Practise Supervised Learning

215 interview questions in this topic.

Related questions