1
$\begingroup$

When applying the MSE-F (F-test, McCracken, 2007), there are two different values to search for the critical value: $k_2$ and $\pi$. I was told that $k_2$ equals the number of parameters included in the regression.

How do I calculate $\pi$? Is it a ratio of the size of the estimation and evaluation periods?

$\endgroup$

1 Answer 1

1
$\begingroup$

If I am not mistaking, reading McCracken (2007) carefully suggests that

"$k_2$ is the number of excess parameters" and "The second parameter is $π$, which affects the range of integration"

(M.W. McCracken / Journal of Econometrics 140 (2007) 719-752, p.728)

In other words, $k_2$ is your amount of predictors and $π$ is the ratio $(T-w/w)$, where $w$ is a fixed window size used to generate the forecast and $T$ is your time periods.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.