I went looking for a copy of my working paper from the 2013 Joint Statistical Meetings and realized, it does not seem to be online. And neither does my poster! So I figured I should fix that. Here’s a link to the poster and the proceedings paper.
The key idea here comes from my work on flood insurance and trying to understand the county-level effects of flood mitigation grants. Some grants are given to state agencies. In other words, the granularity of reality is insufficient for my goals. Normally, it’s just a data problem. So I investigated, briefly, four logical methods of allocating those grants to local-level jurisdictions to understand the implications of each.