Skip to content

Commit

Permalink
#10 Minor edits to statalist post
Browse files Browse the repository at this point in the history
  • Loading branch information
Jerray committed Jun 2, 2022
1 parent 61e0ba5 commit 9a03389
Showing 1 changed file with 2 additions and 2 deletions.
4 changes: 2 additions & 2 deletions doc/multe-statalist.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@ If your research setting has multiple treatments and a saturated group control w

**multe** can help you measure and avoid contamination bias in such settings.

**Avoids contamination bias:** Based on Goldsmith-Pinkham, Hull, and Kolesár (2022), the **multe** command computes three types of weighted-average treatment effects that avoid contamination bias. These are the equal-weighted averages (i.e. average treatment effect), treatment-specific variance-weighted averages (as in Angrist et al. (1998)), and comparably efficiently-weighted averages (as in Goldsmith-Pinkham et al. (2022)).
**Avoids contamination bias:** Based on Goldsmith-Pinkham, Hull, and Kolesár (2022), the **multe** command computes three types of weighted-average treatment effects that avoid contamination bias. These are the equal-weighted averages (i.e. average treatment effect), treatment-specific variance-weighted averages (as in Angrist et al. (1998)), and comparable efficiently-weighted averages (as in Goldsmith-Pinkham et al. (2022)).

**Measures contamination bias:** Our command also measures the extent to which contamination bias affects treatment effect estimates based on a partially linear model (i.e. the coefficients on X when regressing Y on X and W, given an outcome Y, a vector of treatment indicators X, and a vector of saturated group indicators W). It decomposes such treatment effect estimates into own-effect and contamination bias terms.

For an example of usage, see the Stata helpfile. Example data from Project STAR are available for download directly with the package (via `ssc install multe, all`) or through our online repository [here](https://raw.githubusercontent.com/gphk-metrics/stata-multe/ab353845e9cc4d3f30563c345342daff2ee1dec8/test/example_star.dta).
For an example of usage, see the Stata helpfile. Example data from Project STAR are available for download directly with the package (via `ssc install multe, all`).

0 comments on commit 9a03389

Please sign in to comment.