Skip to content

FRM Part I · FRM Exam Part I · Regression Diagnostics

In a regression with n = 50 observations and k = 4 explanatory variables (plus an intercept), one observation has leverage h = 0.30. Using the common rule of thumb that flags leverage above 2(k+1)/n, which statement is correct?

The threshold is 2(k+1)/n = 2×5/50 = 0.20. The observation's leverage of 0.30 exceeds it, so it is flagged as high leverage. Using k rather than k+1 would give a wrong threshold of 0.16, even though it would reach the same flag.

  1. AThe threshold is 0.10 and the observation is flagged as high leverage
  2. BThe threshold is 0.16 and the observation is not flagged
  3. CThe threshold is 0.20 and the observation is flagged as high leverageCorrect
  4. DThe threshold is 0.20 and the observation is not flagged

Explanation

Threshold = 2(4+1)/50 = 10/50 = 0.20. Since 0.30 exceeds 0.20, the observation is flagged as high leverage. Using k instead of k+1 gives 0.16, which would also flag it, but the stated value would be wrong.

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