3 ms·
You'll run out of data pretty quickly. https://statmodeling.stat.columbia.edu/2018/03/15/need-16-times-sample-size-estimate-interaction-estimate-main-effect/ h
by aseerdbnarng 6y ago
You'll run out of data pretty quickly.
https://statmodeling.stat.columbia.edu/2018/03/15/need-16-times-sample-size-estimate-interaction-estimate-main-effect/ https://statmodeling.stat.columbia.edu/2018/03/15/need-16-ti...
- alisonkisk 6y agoHierarchial model has the same problem, since it splits up the data into categories.