OLS · Diagnostics
Heteroskedasticity
When residual spread changes across the fitted outcome, conventional standard errors may be unreliable.
Page 1 of 7 · Big Picture
The big picture: Heteroskedasticity
Does outcome variability change across fitted values?
Why a policy analyst should care
Changing residual variance may leave OLS coefficients centered correctly while making conventional standard errors and confidence intervals unreliable.
Follow the idea from the opening question to a practical decision. These are the conceptual moves this module will develop:
1 Question Does outcome variability change across fitted values?
2 Explore Read a fan-shaped residual pattern
3 Formalize Constant variance is an inference assumption
4 Transfer Audit changing variance
Use the left and right arrow keys to move between pages.