Skip to content

FRM Part I · FRM Exam Part I · Nonstationary Time Series

An analyst fits an AR(1) model to monthly changes in a bond spread over 2005-2023 and suspects that the intercept and slope changed after a regulatory reform in 2012. Which procedure is most directly designed to test for a structural break at a known date?

The Chow test is the right choice. It compares a pooled regression with separate regressions before and after the known break date, using an F-statistic on residual sums of squares. Unit root, autocorrelation and normality tests do not test for parameter changes at a specific date.

  1. AAugmented Dickey-Fuller test on the full sample
  2. BChow test comparing restricted and unrestricted residual sums of squaresCorrect
  3. CDurbin-Watson test on the full-sample residuals
  4. DJarque-Bera test on the full-sample residuals

Explanation

The Chow test splits the sample at the known break date and compares the pooled (restricted) fit with separate sub-sample (unrestricted) fits using an F-statistic. The ADF test checks for a unit root, not a parameter change. Durbin-Watson tests serial correlation and Jarque-Bera tests normality, so neither targets a break at a known date.

Did you get it right without looking?

One question tells you little. A timed set on Nonstationary Time Series shows your real accuracy, how long you take and where you lose marks.

Practise Nonstationary Time SeriesFree diagnostic mock

More Nonstationary Time Series questions