FRM Part I · FRM Exam Part I · Regression Diagnostics
A regression of quarterly sales growth on an interest rate variable produces residuals with a first-order sample autocorrelation of 0.30 over a long sample. Using the approximation DW = 2(1 - rho), what is the Durbin-Watson statistic?
The Durbin-Watson statistic is approximately 2 times (1 minus the autocorrelation). With rho of 0.30 this gives 2 x 0.70 = 1.40, which is below 2 and indicates mild positive serial correlation in the residuals.
- A0.60
- B1.40Correct
- C1.70
- D2.60
Explanation
DW = 2(1 - 0.30) = 2 x 0.70 = 1.40. Option A forgets to subtract from 1 and doubles only... actually it uses 2 x 0.30. Option D uses 2(1 + 0.30), which has the wrong sign. Option C is 2 - 0.30, which omits doubling rho.
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 a stock's return on three factors uses 64 monthly observations. The total sum of squares is 400 and the residual sum of squa…
- An analyst fits a linear regression of monthly fund returns on a market factor and plots the residuals against the fitted values. The plot s…
- A multiple regression with 5 slope coefficients and 56 observations is tested for joint significance of all slopes. The explained sum of squ…
- In the true model Y = 2 + 1.5·X1 − 0.90·X2 + e, the regressors X1 and X2 have correlation 0.50, Var(X1) = 4 and Var(X2) = 9. An analyst omit…
- In a simple regression with n = 25 observations, one observation has a standardized (studentized) residual of 3.4, but its leverage is very …
- Which approach is a standard remedy when a regression's residuals exhibit serial correlation and the analyst wants valid inference without c…