I am working with a complex survey design, with bootstrap weights. I have 1000 bootstrap weights per observation.
My code would be the following:
Code:
svyset [pweight=wgt] bsrweight (bsr001-bsr1000) bsn(?) vce(bootstrap) dof(1000) mse
-Dof is set to 1000 because of the # of replications. (is this correct?)
My question is what I should set bsn too. I am a bit confused about how it works and don't find the stata manuals all that clear. Could anyone help me out?
It says "should have the value of the number of bootstrap samples used to produce each mean bootstrap weight" in the User Guide. how do I do this in Stata?
0 Response to bsn in svyset foor bootstrap regression
Post a Comment