Skip to content

FRM Part I · FRM Exam Part I · Regression Diagnostics

An analyst regresses monthly excess returns of a fund on the market excess return and plots the residuals against the fitted values. The plot shows a clear funnel shape, with residual dispersion widening as fitted values increase. Which conclusion is most appropriate?

A funnel-shaped residual plot signals heteroskedasticity. OLS coefficients remain unbiased and consistent, but the conventional standard errors are incorrect, making t-tests and confidence intervals unreliable. Robust standard errors, such as White's, are the usual remedy. It does not indicate bias, serial correlation or perfect collinearity.

  1. AThe errors are heteroskedastic, so the OLS coefficient estimates remain unbiased but the usual standard errors are unreliableCorrect
  2. BThe errors are serially correlated, so the OLS coefficient estimates are biased
  3. CThe regressors are perfectly collinear, so the model cannot be estimated
  4. DThe residuals are normally distributed, so no correction is needed

Explanation

A funnel pattern in residuals versus fitted values indicates non-constant error variance (heteroskedasticity). OLS slope estimates stay unbiased and consistent, but conventional standard errors are wrong, so t-tests are invalid. Serial correlation would show in residuals ordered over time, not a funnel shape.

Did you get it right without looking?

One question tells you little. A timed set on Regression Diagnostics shows your real accuracy, how long you take and where you lose marks.

More Regression Diagnostics questions