Skip to content

FRM Part I · FRM Exam Part I · Regression Diagnostics

A risk analyst estimates a regression of daily portfolio returns on a factor and finds residuals with a first-order autocorrelation of 0.50. The regression includes no lagged dependent variable. Which statement about the consequences and remedy is most accurate?

With autocorrelated residuals and no lagged dependent variable, OLS coefficients remain unbiased but are inefficient and the conventional standard errors are wrong. Newey-West (HAC) standard errors fix inference by allowing for both autocorrelation and heteroskedasticity.

  1. AOLS slopes are biased and inconsistent; the remedy is to drop the intercept
  2. BOLS slopes remain unbiased but are inefficient and standard errors are invalid; Newey-West (HAC) standard errors are an appropriate remedyCorrect
  3. COLS slopes are unbiased and efficient; no action is needed because only R² is affected
  4. DOLS slopes are biased downward; the remedy is White heteroskedasticity-robust standard errors

Explanation

Without a lagged dependent variable, autocorrelated errors leave OLS unbiased and consistent but inefficient, and the usual standard errors are invalid. HAC (Newey-West) standard errors correct inference. White errors address only heteroskedasticity, not autocorrelation.

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.

Practise Regression DiagnosticsFree diagnostic mock

More Regression Diagnostics questions