Sample size calculation (general and relation to mixed linear model)

[Jakob Nielsen]

Dear Statalisters

I have a question that I hope you can help me with.

I have searched the forums, the internet and available text books and have come up short - I use STATA/IC 16.0.

Basically, I would like to perform a sample size calculation specific for a mixed model with repeated measures (w. correlation between repeated measures).

Specifically, I would like a to reproduce something like the power repeated function specific for ANOVAs described in the STATA guide (https://www.stata.com/manuals13/psspowerrepeated.pdf) - just for a mixed linear model.

First and foremost, any suggestions for a STATA function able to calculate a sample size specific for a mixed model with repeated measures is highly appreciated. Preferably with a function that does not require a prediction of error variance, as I do not have data to support a precise error variance.

If a possibility for a similar calculation specific for mixed model does not exist in STATA - any suggestion of existing STATA functions in addition to the ANOVA function described in the link above is also appreciated.

Thanks in advance, Jakob