My task is to show that Fisher's criterion distribution when the null hypothesis is true follows $F(k-m,N-k)$ when df increases. Here $k-m$ is number of restriction.

So I can write a null hypothesis H0: $b_1=b_2=\ldots b_k=0$.
And I think that I can show that F-statistic follows $F(k-m,N-k)$ using restricted and unrestricted models. where $F=R^2u-R^2r/(1-R^2u) (k-m/N-k)$ or I can find this value from the table of Wald test.

When I compare it critical value =@qfdist(.95,eq.@ncoef-eq2.@ncoef,eq.@regobs-eq.@ncoef).

Is my thinking right? How use should use the condition of task that degrees of freedom increases?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.