The data i created contains heteroscedasticity. I already calculated the power so my idea was to basically do the same but switch the hypothesis so that H0: Heteroscedasticity and H1: Homoscedasticity and then count how many times the new H0 is rejected. But this is not possible in the bptest() feature i use in the lm-package. Can anyone help me? im completely new to R so i am struggling... is there any other way to calculate power of a test or switch the hypothesis?
Asked
Active
Viewed 19 times
0
-
I don't think that would work. In the Breusch-Pagan LM test, $H_{0}$ is an equality (all residuals are distributed with equal variance) while $H_{1}$ is not. Maybe I'm wrong but I don't think you could calculate the sampling distribution under $H_{1}$ (your "new" $H_{0}$ in your inverted test). – Durden May 24 '23 at 21:43