Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: ATE with binary outcome

From   Ronnie Babigumira <>
Subject   st: ATE with binary outcome
Date   Sun, 27 May 2007 21:56:59 +0200

Dear Listers

(Setup: Stata 9.2, WinXP, -pscore- version 2.02)

I'm a novice at the matching game so I beg your indulgence if I ask the obvious.

I am trying to use propensity score matching to estimate the average treatment effect (ATE). However, my outcome is binary (0,1).

My plan was to use -pscore- [SJ-2-4, Becker & Ichino] and following the example in the article,

a. run -pscore- to estimate the propensity score
b. run one of the att* programs to estimate the ATE

Here are my questions

1. Like most of the other work I have seen, the example deals with a continuous outcome. Can I extend this to binary outcome (if yes, would anyone have an example of publication that has gone down this road)?

2. If it is OK (I just tried it out with my data), what would be the interpretation of the result below

ATT estimation with Nearest Neighbor Matching method
(random draw version)
Analytical standard errors

n. treat. n. contr. ATT Std. Err. t

18893 9532 -0.348 0.007 -48.286

Note: the numbers of treated and controls refer to actual
nearest neighbour matches

Many thanks



SJ-2-4 st0026 . Est. of average treatment effects based on propensity scores
(help pscore, attnd, attr, ... if installed) S. O. Becker & A. Ichino
Q4/02 SJ 2(4):358--377
propensity score matching estimators
* For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index