APBREAKTEST—General test for breaks in linear regression

[TomDoan]

Sounds like you're running a model with quite a few lags. It's possible that the pi option is too small given the size of the data set and size of the model you're running.

[Osabuohien247]

Sounds like you're running a model with quite a few lags. It's possible that the pi option is too small given the size of the data set and size of the model you're running.

Thank you Tom, i have figured-out what the problem was. It's ok now.

[Osabuohien247]

Good day everybody,

When the residual dates in Andrews-Quandt and Andrews-Ploberger comes before the Variables (All Coeff) structural break point dates (i.e the robust F-statistics date (1981) before Break point statistics date (1987)),Please what does this denote?

Thank you.

OO

[TomDoan]

I'm not sure what your question is. @APBREAKTEST with ROBUSTERRORS uses a different test statistic than @APBREAKTEST without it so there is no reason that the two would show the maximum break at the same location.