Propensity Scores, R Packages, and Practical Advice with Noah Greifer | Season 6 Episode 3
Apr 10, 2025•1 hr 22 min•Season 6Ep. 62
Listen to this episode in Metacast mobile app
Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more
Summary
In this episode, Noah Greifer discusses propensity score weighting, R packages like WeightIt and MatchIt, and practical advice for causal inference, including targeting different populations and interpreting balance. The conversation covers the nuances of overlap weights, balance diagnostics, M-estimation, and the importance of balancing statistical rigor with audience understanding. Noah shares insights on numerical differentiation, chain rules, and the value of making statistical methods accessible through R packages.Episode description
Noah Greifer is a statistical consultant and programmer at Harvard University.
Episode notes:
- WeightIt package: https://ngreifer.github.io/WeightIt/
- MatchIt package: https://kosukeimai.github.io/MatchIt/
- Noah's awesome Stack Exchange post: https://stats.stackexchange.com/a/544958
Follow along on Bluesky:
- Noah: @noahgreifer.bsky.social
- Ellie: @EpiEllie.bsky.social
- Lucy: @LucyStats.bsky.social
🎶 Our intro/outro music is courtesy of Joseph McDade. Edited by Cameron Bopp.
For the best experience, listen in Metacast app for iOS or Android