Closed henry940926 closed 1 year ago
Many thanks for your message. SKAT p-values are currently obtained through permutation procedures. This makes it difficult to integrate our SKAT extension into SKAT-O which tests for different contributions of SKAT and burden tests. In addition, when SKAT-O is significant, it is often mainly driven by either the burden or the SKAT test. Nevertheless, this question is of great interest and I would be happy to further talk about it.
Best, Ozvan
Dear author,
I came across your papers (the extension of burden test and SKAT to multinomial outcomes) which are amazing ideas. I wonder if you have thought/are in the process of extending the SKAT-O method to a multinomial outcome?
Many thanks