[R] GenMatch providing different results
Hi there, I'm performing some propensity score matching in R, where I am using GenMatch() to generate a weight matrix for the covariates, before plugging that into Match(). However, I keep on getting different results each time I run the code. I'm using int.seed and unif.seed in GenMatch, but to no avail. I've also noticed that when I just use Match() without previously using GenMatch(), the results are consistent each time, which is why I believe GenMatch() is causing the problem: A slight extract of my code looks like this: ##Genetic Matching gmatch <- GenMatch(Tr=treatment, X=X, estimand="ATT", M=1, pop.size=50, max.generations=15, wait.generations=5, caliper=0.05, ties=FALSE, int.seed=240690, unif.seed=021189) ###Actual matching match <- Match(Y=outcome, Tr=treatment, X=X, estimand="ATT", Weight.matrix=gmatch) summary(match) Can anyone point me in the right direction to diagnose my problem? Thanks, Rob. -- View this message in context: http://r.789695.n4.nabble.com/GenMatch-providing-different-results-tp4700836.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] P-value from Matching
Hi all, When using the match command from the matching package, the output reports the treatment effect, standard error, t-statistic and a p-value. Which test is used to generate this p-value, or how us it generated? Thanks, Rob. -- View this message in context: http://r.789695.n4.nabble.com/P-value-from-Matching-tp4703309.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] P-value from Matching
Hi Peter, That was my first port of call before I posted this thread. Unfortunately, it does not seem to explicitly state which test is used or how the p-value is calculated. Thanks, Rob. -- View this message in context: http://r.789695.n4.nabble.com/P-value-from-Matching-tp4703309p4703345.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
