A Bayesian Multilevel Reading List
Here’s a list of readings on Bayesian Hierarchical Models with specific reference to multilevel modeling. This is a work in progress but it would work reasonably well as the basic reading list for an advanced topics course in Bayesian multilevel modeling….
Integration for Nonlinear Models with Lots of Random Effects
Over the last few years I’ve been getting more interested in how to deal with high dimensional integration of random effects. In straightforward nonlinear mixed effects models¹ you typically try to approximate a random effect by numerical integration. The standard…
Books on Multilevel, Longitudinal, and Panel Analysis
Here lies my current list of books on multilevel, longitudinal, and panel data modeling. I’ve broken them down into rough categories but there’s a lot of heterogeneity within groups here. I’m explicitly avoiding including articles on this list because there are…
I’m Teaching at ICPSR!
Big news! The ICPSR summer program is a social science statistics summer camp housed at the University of Michigan. The courses are mostly for grad students (and some faculty and even the occasional undergrad) designed to help provide training that…
Stata or R for Multilevel Modeling
Like most people in political science I was initially trained on Stata. It tends to be the default software in most of the social sciences and public policy. R is usually used by statisticians and the various flavors of methodologists…
Nonlinear FE: Incidental Parameters Bias
Probably the most common technique for dealing with repeated observations in economics and a lot of the social sciences is to segregate between group and within group variability and only look at the within group effects of explanatory variables on…
Discrete Choice Methods with Simulation (Nonlinear Random Effects Models)
One of the things that had the biggest impact on my early development as a methodologist was reading Kenneth Train’s book Discrete Choice Methods with Simulation. It’s available for free as a PDF on his website. You should read it. Really. The focus…
Fixed, Mixed, and Random Effects: The RE assumptions debate part II
Last week I posted about a question that was sent out on the r mixed effects listserv comparing the different approaches between economists and statisticians to dealing with mixed effects problems. Below is a slightly edited version of the conversation that…
Fixed, Mixed, and Random Effects: The RE assumptions debate
Not long ago a question was posted to the r mixed effects models listserv that focused on the key difference between economists and almost everyone else in dealing with multilevel problems. The question is something that I’ve been grappling with since…
Random Coefficients
One of the key advantages of multilevel models over standard GLMs is the idea of a random coefficient. Random coefficient models are essentially interactions between the random intercept (read latent variable) and a lower level variable so that the effect of…