Kenward-Roger V2 (KR2, 2009)

[Json-G-nomics]

Hello,

I am reaching out to see if there is any plans for integration of the Kenward-Roger degrees of freedom approximation from Kenward and Roger's updated paper.

Michael G. Kenward, James H. Roger,
An improved approximation to the precision of fixed effects from restricted maximum likelihood,
Computational Statistics & Data Analysis,
Volume 53, Issue 7,
2009,
Pages 2583-2595,
ISSN 0167-9473,
https://doi.org/10.1016/j.csda.2008.12.013

It is a refinement of the original KR method, especially for models with intrinsically linear covariance parameterizations. This single implementation has kept me from using R for LMM with repeated measures.

Is there a roadmap for including this new method in MMRM?

Thanks,

Jason

[danielinteractive]

Hi Jason,

Thanks for reaching out!

We have not coded KR2 so far, just because we have KR already and it is much more complex.

That said, happy to discuss if funding is available

Cheers
Daniel