I did something similar once. Here is an example of the way I did it,
given pre-constructed weights
set to zero for those not in the (sub)sample (i.e. bootstrap or
jacknike..). There is no need to use the canned programs to do this.
bill magee
#delimit ;
set more off;
gen bagespline1=0;
gen bagespline2=0;
gen bagespline3=0;
gen bagespline4=0;
gen brecentim=0;
gen bltim=0;
I'd like to run analyses using jackknife estimation. My dataset has a
frequency weight and replicate weights (50) to be used in the jknife
command. Stata does not seem to allow for specifying use of these
weights.
Can anyone help me with the syntax, or does Stata really not allow that?
Also, is there a way to get Stata to allow the use of non-integer
frequency
weights?