In the social sciences we talk a lot about groups being different from one another. We might also talk about things like spending on certain programs going up or down over time.
I previously wrote about a project in which I was attempting to figure out how to build a probability density function for a hurdle log-normal model. Ultimately I kind of left this topic hanging because I wasn’t really sure that the solution I had settled on was correct.
As a part of a larger project I’ve been working with BRMS and some models and family functions that are new to me. In particular, my work on troop deployments has made me think more about hurdle models.