search

QuickPay-Operational-Performance / Data-and-code

Data and code for econometric analysis
0 stars 0 forks source link

Difference-in-difference with matching #59

Closed JNing0 closed 3 years ago

JNing0 commented 4 years ago

I followed up Vlad's earlier comment and looked up diff-in-diff with matching. Cui et al used the matching procedure in this highly cited methodological paper on matching. The authors of the matching paper developed an R package called MatchIt (see details in the paper). It might be a good idea to use their package when we do matching for robustness.

vob2 commented 4 years ago

I am no expert, but I believe most statistical software tools come with some kind of matching packages. In R it is MatchIt. In Stata they are psmatch2 to effects (the latter is apparently newer).

vob2 commented 4 years ago

psmatch2 and teffects

Autocorrect change the name of the package from teffects to effect

vibhuti6 commented 4 years ago

Thanks Jie and Vlad. I have the code for matching the USAspending data using the MatchIt package. So hopefully this would be straightforward to implement.

vibhuti6 commented 3 years ago

Addressed in another issue, see here.