Dear all,

I am getting new with matching firms.
The objective of my analysis is to perform a DiffinDiff looking whether a shock that happened in 2018 Europe had an effect of Y variable of interest.
Given that the shock affected all EU firms, I had to find a proper control gorup. So I decided to match EU firms (treated) with US firms (control group).

Here is where I'm a bit stuck. I know that there are different types of matching procedure (psmatch2, caliper match, teffects nnmatch). Nevertheless I'm not able to figure it out which one is more suitable for my purpose.

Indeed, there are some variables which I need an exact match, such as industry (SIC) and year_quarter (datafqtr). Others they don't need to be exactly matched such as assets (lnassets), profitability (roa), market to book (mtb).

Below you will find an example:

Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input long gvkey float(lnatq roa mtb datafqtr) int sic float treated
1004  7.412155    .011152683  1.347024 202 5080 0
1004  7.440574     .01275107  1.255999 203 5080 0
1004  7.468726    .009606206 1.1233281 204 5080 0
1004  7.507477     .00983832  .8514237 205 5080 0
1004  7.705394    .010224387  .9980568 206 5080 0
1004  7.694235    .005832952 .56125623 207 5080 0
1004  7.682621    .008337241  .6751086 208 5080 0
1004  7.691337    .008165887  .6753602 209 5080 0
1004   7.70391    .008351679   .755529 210 5080 0
1004  7.667111  .00027561496  .8600075 211 5080 0
1004  7.663502     .00839174 1.0584167 212 5080 0
1004  7.675963    .008167837  1.279561 213 5080 0
1004  7.702149    .007736324 1.1586059 214 5080 0
1004  7.695985    .006753388  .9617889 215 5080 0
1004  7.698664     .00535767  1.100439 216 5080 0
1004  7.708725    .006271007 1.0185193 217 5080 0
1004   7.66566   .0008712599 1.2441177 218 5080 0
1004  7.323171     -.0450011 1.2381912 219 5080 0
1004  7.344202    .004833126  .9994889 220 5080 0
1004  7.336025    .005970537  .9978906 221 5080 0
1004  7.327781    .006477786  .8619188 222 5080 0
1004  7.273856    .007961945  .9731014 223 5080 0
1004   7.26333    .006901118  .9796414 224 5080 0
1004  7.304247    .007825371  1.451624 225 5080 0
1004   7.31462    .009636297 1.3338964 226 5080 0
1004   7.31595    .010844256 1.3129827 227 5080 0
1004  7.334134    .007246854 1.3501345 228 5080 0
1004  7.342326    .008647595  1.592975 229 5080 0
1004  7.321321     .02048094 1.6115916 230 5080 0
1004  7.329553     .01192005 1.6570096 231 5080 0
1004  7.338108    .012342857 1.7601588 232 5080 0
1004  7.380381    .007129217  1.638842 233 5080 0
1004   7.34375    .017392961   1.42428 234 5080 0
1004  7.324622    .017364625 1.1555039 235 5080 0
1004  7.428392    .010686498  1.651799 236 5080 0
1004  7.470167    .011692844 1.6928574 237 5080 0
1004   7.51147   .0014509738  1.304863 238 5080 0
1004  7.639642   -.007676757   .784297 239 5080 0
1004  7.446468   -.007329677  .7998812 240 5080 0
1004   7.42016    .008512902 1.1116177 241 5080 0
1004  7.404157    .018779626  1.506853 242 5080 0
1045 10.130145    -.00388542 -.6584475 203 4512 0
1045 10.686727    .013576821 18.508453 219 4512 0
1045 10.787565    .067558944  4.694364 223 4512 0
1045  10.84494    .005645524  6.257743 227 4512 0
1045 10.847316   -.011233466  6.301753 231 4512 0
1045  11.01172    .005723623 -87.51609 235 4512 0
1045 11.002016    .006833375 -104.0751 239 4512 0
1045  11.03502   -.034909163 -1.427223 243 4512 0
1050 4.2912674    .009291923 2.5145016 202 3564 0
1050 4.3146977    .010659452  2.426259 203 3564 0
1050 4.3039703     .01691042 2.3047729 204 3564 0
1050 4.2912674     .02656203 2.4957125 205 3564 0
1050 4.3080707    .031474795 2.0078619 206 3564 0
1050 4.3738055    .036214057 1.8740126 207 3564 0
1050  4.364995     .02588706  2.503418 208 3564 0
1050  4.356016    .031737924  2.463261 209 3564 0
1050 4.4362187     .04024606 2.8538914 210 3564 0
1050 4.5444007     .03418478  2.721909 211 3564 0
1050 4.7370315    .021210477 3.2661114 212 3564 0
1050 4.7669487     .02627183 2.9947255 213 3564 0
1050  5.868268   -.006201241 2.1383796 214 3564 0
1050  5.853742     .00788126  2.425274 215 3564 0
1050  5.816536    .008828905  2.473133 216 3564 0
1050   5.81002      .0134233 2.2780898 217 3564 0
1050  5.859509     .01082467  1.938308 218 3564 0
1050  6.026747    .004863273 2.2524014 219 3564 0
1050  6.034112     .00047608 1.5544752 220 3564 0
1050  6.032465    .005044541 1.6500274 221 3564 0
1050  6.472068   -.009073292 1.1353052 222 3564 0
1050  6.394959   -.004943647   1.08922 223 3564 0
1050  6.366295    .005251046  .8753731 224 3564 0
1050  6.383003     .00690184 1.2307675 225 3564 0
1050  6.331427     .01006189 1.5693275 226 3564 0
1050  6.211873    -.09649704  2.507128 227 3564 0
1050  6.178575  .00007747686   1.91266 228 3564 0
1050  6.127655      .0116641  1.626823 229 3564 0
1050   6.12149    .006644031 1.4836886 230 3564 0
1050  6.083472    -.02592351   .950555 231 3564 0
1050  6.045287    .013391923  .7913911 232 3564 0
1050  6.043585  -.0021362859 1.1158966 233 3564 0
1050  6.004427    -.03124773 1.5463287 234 3564 0
1050  5.972745    .002336422  1.316129 235 3564 0
1050  5.983102    .004723466 1.4141856 236 3564 0
1050  5.970562     .01399045 1.8406892 237 3564 0
1050  5.965223    .004942626 1.3314334 238 3564 0
1050  6.012827     .02103781 1.3944337 239 3564 0
1050  6.098651    .007991624  .8447016 240 3564 0
1050  6.036419    .007544732 1.1705873 241 3564 0
1050  6.045218 -.00056872127 1.2851752 242 3564 0
1050   6.03862   .0042310236  1.214629 243 3564 0
1072  7.690894     .03175445   1.22638 201 3670 0
1072   7.70624      .0274581  1.339759 202 3670 0
1072  7.749099    .027857693 1.2438934 203 3670 0
1072  7.781409    .028673224  1.231708 204 3670 0
1072  7.775674     .02592056   .943126 205 3670 0
1072   7.77904    .015453297 1.0126293 206 3670 0
1072  7.811168   -.005592435 1.0604296 207 3670 0
1072  7.847062    -.05442819  .9279667 208 3670 0
1072  7.853507     .01092431  .8205881 209 3670 0
end
format %tq datafqtr
For the moment, I created an indicator (treated) whether the firms is either in Europe (equals to 1) or in US (equals to 0). Moreover, I guess (feel free to correct me if I'm wrong) that the matching should be based on characheristics before the shock. Therefore I created another indicator (match_period) identifying a matching period (ideally from 2016q1 to 2017q4)


Would you be so kind to give me some advices on that?. I know that psmatch2 could be the right

I also searched some example already discussed on this forum but without really getting it.