FRM Part I · FRM Exam Part I · Regression Diagnostics
A regression of bond spread changes on equity returns shows a clear U-shaped pattern in the residuals when plotted against the fitted values. Which is the most appropriate interpretation and response?
The U-shaped residual pattern indicates a misspecified functional form, meaning the true relationship is nonlinear. The appropriate response is to add a squared term or transform the variables, rather than treating the problem as multicollinearity or ignoring it.
- AHeteroskedasticity is the only issue; use a larger sample and keep the model
- BThe pattern suggests a misspecified functional form; consider adding a squared term or transforming the variablesCorrect
- CMulticollinearity is present; drop the equity return variable
- DThe residuals are normally distributed; no action is needed
Explanation
A systematic curved pattern in residuals versus fitted values indicates a nonlinear relationship the linear model fails to capture, which is a functional form misspecification. Adding a quadratic term or transforming variables addresses it. Multicollinearity concerns correlation among regressors, not residual patterns.
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
- In a multiple regression of monthly fund excess returns on three risk factors, an analyst finds that two of the explanatory variables have a…
- Two models explain bond spread changes using n = 54 observations. Model A has 3 explanatory variables and R-squared of 0.60. Model B adds 2 …
- An analyst regresses monthly excess returns of a fund on a market factor using 120 observations. The Durbin-Watson statistic for the residua…
- An analyst regresses a fund's returns on two highly correlated factors. The standard error of the coefficient on factor 1 is 0.30 when facto…
- 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, regressing X2 on X1 gives a …
- In testing residuals for serial correlation, a Durbin-Watson statistic is computed as 2.90 for a regression with a sufficiently large sample…