Modeling Ordinal Change over Time
Say that you want to model ordinal change in status (e.g. a Likert scale) over time between two or three time periods. How do you go about it? I had to put a decent amount of thought into this problem…
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…
Independence across Levels in Mixed Effects Models
One issue that comes up every so often is the appropriate number of levels to have in an analysis. Any fixed or random effects model has at least two levels but you can presumably come up with more in most applications. This…
Nonlinear Multilevel Modeling is Hard
Often when people talk about multilevel modeling what they really mean is HLM (hierarchical linear modeling). This is a multilevel model for a linear(ish) and continuous(ish) dependent variable and it amounts to a relatively trivial complication for a GLS model…