Bookmark and Share

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

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

st: Pscore command

From   Nyasha Tirivayi <>
Subject   st: Pscore command
Date   Fri, 12 Nov 2010 12:46:57 +0100


I used the pscore command for propensity score matching. The balancing
property was satisfied after the model had 14 covariates (from a start
point of 20, pasmatch2 also balances with less covariates). Someone
has said these covariates are not enough to satisfy the conditional
independence assumption. Is there a minimum number of covariates
needed in propensity score matching?

Kindly advise

*   For searches and help try:

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