I am doing post-hoc comparisons of contrasts based on linear mixed models I built in R. I am using the emmeans package for the comparisons. One of the default adjustment methods for multiple correction is "mvt" which stands for "multivariate $t$". However, I cannot find a good explanantion of this method. Is there a paper available I can cite?

  • $\begingroup$ I'd recommend expanding the question a little. Say, for example, what mvt stands for (it can be found in the documentation/vignettes and help files of the package). Include the explanations that the authors offers (see here, for example) as a quote so that all readers are on the same page. Don't expect that readers here will google the terms and the package on their own. $\endgroup$ – COOLSerdash Dec 13 '18 at 9:27
  • $\begingroup$ Simply put, the method estimates a multivariate t distribution of all your hypotheses which it uses to generate p-values. The multivariate approach takes into consideration the correlation of the hypotheses. $\endgroup$ – user2974951 Dec 14 '18 at 8:02

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.