Hi, I am working with Twang. I am running the specification suggested by RAND tutorials for binary treatment in STATA, and I do not know how to save the propensity score. In other words, the specification suggested on the tutorial returns a set of files and datasets. The STATA dataset includes weights but it does not include the propensity score (or at least I have not found it). Since I am estimating ATT effects, I can not infer the ps from weights for the group of treated. In this context, does anyone know how to save the propensity scores?
Suggested specification:
ps treat x1 x2 x3 x4, ///
ntrees(3000) intdepth(2) shrinkage(0.01) ///
permtestiters(0) stopmethod(es.mean) estimand(ATT) ///
rcmd("$Rlocation") ///
objpath("$folder") ///
plotname("$folder\ex1.pdf")
Thanks,
Related Posts with Twang - ps
Help with reshaping dataHi Statalist, I'm having trouble with reshaping my data. The data has a (self-explanatory) variable…
Calculate and visualize horizontal distance between 2 (time series) linesHello, Is there a convenient way to calculate and visualize horizontal distance between 2 lines? See…
Surpress "percent" label on "x"-axis of graph hbarDear Statalists, I coded a horizontal bar graph which looks exactly how I desire it. Only problem i…
Dummy variablesI have to complete an assigment where I have to assign a dummy variable to 7 different classes. Thos…
Durbin-Wu-Hausman test for GEEsDear Stata Forum I am running a GEE panel regression. I would like to check for endogeneity using …
Subscribe to:
Post Comments (Atom)
0 Response to Twang - ps
Post a Comment