I am trying to test for sorting of individuals across the threshold in an RD setting. In order to make my sample representative I need to apply weights, nevertheless -rddensity- and -rdbwdensity- (the commands I found to be most commonly used for this purpose) don't accept them. Therefore, if I run the commands without specifying weights the density across the threshold is not representative of the true one.
My data looks like:
id | income | factor |
1 | 1149.2 | 3 |
2 | 1029.4 | 4 |
3 | 1345.62 | 6 |
4 | 674.2 | 6 |
5 | 886.36 | 1 |
... | ... | ... |
The command I have on mind would be annalogous to the the following (if weights were allowed):
rddensity income [fw=factor], c(1000) bwselect(each)
My question is: Is there a way I can get around this problem or a well stablished command with similar purpose than -rddensity- but allowing for weights?
Thank you very much in advance for any useful thoughts on this issue.
0 Response to Regression Discontinuity - manipulation test applying fweights problem
Post a Comment