I recently had to review a paper which led me down a rabbit hole. I’ll save you the details, but it prompted me to do a power analysis for a Random Intercept Cross-Lagged Panel model (RI-CLPM). I thought I might as well share my attempt, in case somebody else out there has to do this. Hopefully, this is of use to somebody. I recommend that you also read Mulder, 2022 and note ‘caveat emptor’ - you might want to scrutinise further and also kick the tires with regard to assumptions such as multivariate normality, etc.
You can find my attempt here and this .rmd. As a bonus, you’ll find out what I made of the authors’ suggested sample size, but I guess you can tell from the picture of the post… .
Enjoy!