FRM Part I · FRM Exam Part I · Regression Diagnostics
A risk analyst plots OLS residuals against fitted values and sees a funnel shape, with residuals spreading out as fitted values increase. Which action is most appropriate if the analyst wants valid inference without changing the coefficient estimates?
The analyst should use heteroskedasticity-consistent (robust) standard errors. A funnel-shaped residual plot signals non-constant error variance, and robust standard errors correct inference while leaving the unbiased OLS coefficient estimates unchanged, unlike dropping data or adjusting confidence levels.
- ADrop the largest fitted observations from the sample
- BUse heteroskedasticity-consistent standard errorsCorrect
- CAdd a lagged dependent variable to remove autocorrelation
- DIncrease the confidence level to 99% for all tests
Explanation
The funnel pattern indicates heteroskedasticity. Robust (White) standard errors keep the OLS coefficients but correct the standard errors for valid inference. Dropping observations is arbitrary, the lag addresses autocorrelation, and changing the confidence level does not fix the variance estimate.
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
- A regression of daily portfolio returns on a market factor shows that the variance of residuals is much larger in high-volatility periods th…
- The true model is Y = 1.0 + 2.0*X1 + 3.0*X2 + e. An analyst omits X2 and regresses Y on X1 only. In the sample, Cov(X1, X2) = 0.8 and Var(X1…
- In a multiple regression, the auxiliary regression of explanatory variable X1 on the other explanatory variables yields an R-squared of 0.90…
- A risk analyst's regression of portfolio returns on two highly correlated factors (sample correlation 0.97) is used only to forecast returns…
- An analyst regresses squared residuals from an OLS model on the explanatory variables as part of a Breusch-Pagan test. The auxiliary regress…
- A regression has residual standard error s = 2.0. One observation has residual e = 7.0 and leverage h = 0.19. Using the internally studentiz…