textreg - Concise comparative summarization of text (you can regress a category label onto the full space of words and phrases in a corpus). CRAN R package.
simITS - Package for evaluating Interrupted Time Series data using simulations that have autoregressive structure. This gives nice displays of projected uncertainty that grows as one extrapolates farther and farther form the policy change. CRAN R package.
blkvar - Package for analyzing multisite and blocked experiments in various different ways. R package on GitHub. Install this package in R using
hettx - Package for testing for and estimating treatment variation in individually randomized trials. CRAN R package. Also see GitHub for most recent version.
Random selection of favorite books and papers
The following books and papers are cornerstones of how we think about statistics (more or less). They are certainly worth being familiar with, along with all of our own work, of course.
Lin, W. (2013). Agnostic notes on regression adjustments to experimental data: Reexamining Freedman’s critique. The Annals of Applied Statistics, 7(1), 295–318.
An important investigation of how linear regression works from a randomization viewpoint. It also showcases the Neyman-Rubin causal model, and how finite-sample inference couples with survey sampling techniques.
Rosenbaum, P. R. (2009). Design of Observational Studies. New York: Springer Science & Business Media.
An important overview of causal inference, permutation testing, propensity scores, and matching. More importantly, this book provides much guidance on how to think about evidence in observational studies. See further review here.
Misc Resources and Information
- Insights from JREE: Paper Summaries for the Field. A blog of one page, freely available and written to be accessible summaries of papers published in JREE (part of an initiative started by the current editorial board of which Dr. Miratrix is a part).
- Prof. Luke Miratrix's CV (Click Here.)
- Maxime Richard's award winning poster on geographic regression discontinuity designs.
This work is part of a two-paper series looking at how spatial statistics can be coupled with RDD. For our first paper, see A Nonparametric Bayesian Methodology for Regression Discontinuity Designs. For the second, see A Bayesian Nonparametric Approach to Geographic Regression Discontinuity Designs: Do School Districts Affect NYC House Prices?.
- Miratrix & Weidmann talk: "Using national data and meta-analysis techniques to get a handle on how bad some biases might be in practice". Online Causal Inference Seminar, December 2020