Hi
I have a few questions regarding Propensity Scores Matching with Kernel Matching.
1) If there are 5 outcome measures, do I have to do the matching for separate outcome measures?
2) How to interpret the "t-stat" from psmatch2 Kernel output to determine whether the ATT effect is significant or not?
3) 1:1 or 1:n matching can subset the matched data. Can anyone please advise on how to subset the data after Kernel matching so that can proceed with data analysis on matched data set?
4) Like IPTW, it makes use of all the available data set. Will Kernel matching arrive at the sample size (total, treated and control) ?
5) Kernel matching provides ATT and is there a way to get ATET?
6) In a situation where radius matching and Kernel matching provides the similar reduction in standardized difference of the covariates, which outcome result should we report?
Related Posts with Kernel matching - propensity scores
Haghish's weaver command and wkhtmltopdfDear Stata users, My question is about -weaver- command written by haghish. The problem is that it …
P value for area under the curve using Iroc commandI am using Stata 13. I have run a logistic regression to check the association of Obesity with sever…
Indexing observations for a subject while skipping missing valuesHello Statalisters, I am sure I am missing something obvious but I looked up on the forum and onlin…
Fama MacBeth model - factor loadings: asreg with fmb omits coefficientsDear all, I want to perform Fama and Macbeth (1973) using asreg, but results shows omitted, what is…
Checking cross sectional dependence using XTDCCE2 commandHi I had applied PMG model after the hausman test failed to reject the null hypothesis. However, si…
Subscribe to:
Post Comments (Atom)
0 Response to Kernel matching - propensity scores
Post a Comment