This is very informative. I would note, however, that the three level with covariates example assumes that the three covariates are constructed in such a way that they are mutually uncorrelated which would not be true in actual data. I'm wondering if there is some elegant way to construct these variables with some desired degree of correlation.
↧