Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: no ATET with teffects aipw


From   Sarah Edgington <[email protected]>
To   [email protected]
Subject   st: no ATET with teffects aipw
Date   Mon, 16 Sep 2013 09:44:55 -0700

Hello all,
I have some questions about the new teffects commands, specifically the doubly robust estimator aipw.

My first question is about the documentation for aipw command. Page 62 of the teffects manual makes it clear that the only two statistics that -teffects aipw- can produce are the average treatment effect (ate) and potential-outcome means (pomeans). The average treatment effect among the treated (atet) is not listed as an option. However, in the remarks section of aipw on page 63 it says "They compute the weighted means of the treatment-specific predicted outcomes, where the weights are the inverse-probability weights computed in step 1. The contrasts of these weighted averages provide the estimates of the ATEs. By restricting the computations of the means to the subset of treated subjects, we can obtain the ATETs."

Is this mention of the ATET in the remarks section of the aipw an error? I did try using the atet option with aipw just to double check that it hadn't been excluded erroneously from the options section. Predictably,I got an error message indicating that the atet option is not allowed.

This leads me to my second, more substantitve, question. Why is the ATET not available using -teffects aipw-? As a long time Stata user my assumption is that if a command doesn't allow an option there's a good reason for it. I was wondering if someone could help offer some intuition for what the reason is in the case of aipw?

I know that I can get the atet using the ipwra doubly robust estimator. However, I'd like to also have a clearer understanding of what aipw is doing and why the atet isn't available.

Thanks.

-Sarah
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index