Wald Coefficient Tests

[Steffen]

Hello,

I've got a question to the outcome of some Wald Coefficient Tests.

I estimated a regression and found coefficients of c(1) close to zero and c(2) close to unity. The standard errors are quite small, around 0.003.

When I apply Wald Coefficient Tests for the restrictions c(1)=0 and c(2)=1 the null is overwhelmingly rejected. That surprises me looking at the estimation output. Can anybody explain how estimates so close to the testable restrictions in connection with these small standard errors can lead to such strong rejections of the null?

Many thanks in advance.

[startz]

Small standard errors make rejection more likely.

[Steffen]

Thanks a lot. Obviously I made an error in my reasing.