5 votes

Expected Fisher information isn't positive definite for truncated normal with heteroskedasticity

The expected Fisher information is positive-definite by definition, so there must be some mistake in your code. Here is exactly the same calculation using MATLAB's symbolic math toolbox, which ...
user avatar
  • 3,132
2 votes

Post hoc power analysis for multilevel regression analysis

Answering in reverse: Question 2. First, the examples in the cited paper on power estimates are for large numbers of clusters, each of relatively small size. The plots of minimum detectable effect ...
user avatar
  • 62.4k
1 vote

Multilevel model for education data, if individuals within cluster change (school) between time points

I'm fairly sure this is a classic case of multiple membership, and there are models that have been proposed to account for things like this: https://arxiv.org/abs/1907.04148#:~:text=Multiple%...
user avatar
  • 425
1 vote

How do I interpret the coefficients of a mixed effects multilevel logistic regression differently from regular logistic model?

@Robert Long what would it mean if a variable was significant in the regular logistic regression, but no longer significant after a random effect is added in the mixed-effect model? What does that say ...
user avatar
  • 41
1 vote

Multiple membership model random effects specification

My response is coming many years after the original post. However, in case anyone finds this post in the future, it is possible to modify the lmer function from the lme4 R package to fit multiple ...
user avatar
  • 11

Only top scored, non community-wiki answers of a minimum length are eligible