Interpreting LikelihooD ratio & wald tests?

Question

IT's like I've forgotten everything that statistics ever taught me. But, if I perform a wald test or a likelihood ratio test (I know, they aer not quite the same thing) and I get an ouput of

* Wald test

. test sy2 cy2 // tests the hypothesis that both parameters are zero

( 1) [count]sy2 = 0

( 2) [count]cy2 = 0

chi2( 2) = 30.75

Prob > chi2 = 0.0000

Is it correct to assume that this means that sy2 and cy2 can be dropped from the model becaues they are zero?

Answer 1

No, it means exactly the opposite!  It means at least one of the parameters is clearly not zero.  That probability of 0.0000 is the probability of seeing your data if sy2 and cy2 were zero.  Since this probability is tiny, it's fair to say that these parameters are very unlikely to be zero.

The next stage would be to test each parameter one at a time to see if either of them can be left out.