Closed chstock closed 10 months ago
Thanks for the review, @wlandau! I believe I have addressed all comments and the PR should be ready for another round of review.
Looks great, @chstock. I think this PR is in great shape for a merge. Going forward, we could think about aligning with #72 and discussing https://openpharma.github.io/mmrm/latest-tag/articles/mmrm_review_methods.html.
This is a new vignette with an exemplary comparison of Bayesian and frequentist MMRM fits as suggested in issue #68.
I used the the FEV1 dataset (
fev_data
) from the{mmrm}
package. The endpoint is 'change from baseline in FEV1'. For different reasons, the alternative datasets we had identified appeared not as convincing. We may decide to change the dataset later. Alternatively, we could (later) use a dataset from the new{clindata}
package or a simulated one based on realistic parameters as described e.g. here. However, this does not seem urgent or necessary from my point of view. I adapted the code used in the analysis of theorthodont
dataset (#66).Please let me know what you think.