Andy Timm - My talk on Regularized Raking at NYOSPM

I recently gave a talk on Regularized Raking at the New York Open Statistical Programming Meetup.

Here is the abstract:

Raking is among the most common algorithms for producing survey weights, but it is often opaque what qualities of the resulting weights set are prioritized by the method. This is especially true when practitioners turn to heuristic methods like trimming to improve weights. After reviewing the basic raking algorithm and showing some examples in R, I’ll show that survey weighting can also be understood as an optimization problem, one which allows for explicit regularization. In addition to providing a conceptually crisp view of what (vanilla) raking optimizes for, I’ll show that this regularized raking (implemented via the rsw python package) can allow for more fine-grained control over weights distributions, and ultimately more accurate weighted estimates. Examples will be drawn from US elections surveys.

The slides and reproduction materials can be found here: https://github.com/andytimm/Regularized-Raking. It looks like the presentation on stream froze for a bit in the middle part of the talk, so you may want to pop the slides open to follow along.

For anyone else in the New York area, the meetup is a great group of smart folks working in a bunch of interesting industries- come join us sometime.

The recording is below:

Reuse

https://creativecommons.org/licenses/by/4.0/

Citation

BibTeX citation:

@online{timm2023,
  author = {Timm, Andy},
  title = {My Talk on {Regularized} {Raking} at {NYOSPM}},
  date = {2023-12-05},
  url = {https://andytimm.github.io/posts/NYSOPM_talk_regularized_raking/NYOSPM_talk.html},
  langid = {en}
}

For attribution, please cite this work as:

Timm, Andy. 2023. “My Talk on Regularized Raking at NYOSPM.” December 5, 2023. https://andytimm.github.io/posts/NYSOPM_talk_regularized_raking/NYOSPM_talk.html.