Perhaps it's worth including the link to the paper/code so we know what you're referencing?
1. Please provide sample data - you can find fake data throughout the example questions or in the SASHELP library. PROC PSMATCH which does propensity score matching has some good example data. Only you know the type of data you're going to receive so we make pick something that doesn't align with what you're doing at all.
2. No idea what that means. I could guess you mean a CI for a specific parameter but it's a guess.
3. You should first read some tutorials on macros and write a small one so you understand the basic functionality. Using code you don't understand can be quite dangerous - you'll often miss things. This is a good introductory tutorial to get you started and shouldn't take more than an hour or two to go through.
https://stats.idre.ucla.edu/sas/seminars/sas-macros-introduction/
4. Here's a fully worked macro and example The code to run the macro is at the bottom of the code, you'll need to uncomment the lines - the comments do indicate the test data and how to call it.
https://github.com/Jiangtang/Programming-SAS/blob/master/UserMacros/mayo/gmatch.sas
@mgrzyb1 wrote:
Hi.
I am going to have to use the greedy matching propensity score macro. I am not good at macros. I know Lori Parsons has published a paper.
But I am not sure how to run it. Or where to put the commands to use the macro. I have to figure out the caliper width of 0.2 of the standard deviation of the logit in PS.
1, Can anyone help me with the codes? For example, pretend you have observational data.
2. . How do I calculate the caliper width of 0.2 of the SD of the logit in PS? Where can I find the SD of the logistic regression. I have done tons of logistic regressions, but I do not recall seeing a standard deviation.
3. Even though Lori Parson's paper is wonderful, I do not fully understand macros, where to put the macro call line.
4. If anyone has done this, greedy matching, could you please send me the whole code with where the macro call should be? and a separate SAS macro they used with their variables?
Thank you.
I have no data yet, so for #4, if you could include non-identified data so I can complete this, I would be most thankful.
PLEASE HELP ME!!!
