Calculate average of the (pairwise) sample cross-correlations

[Jonas Petter]

Dear forum,
since this is my first post here and since I am new to Stata, I hope my post fulfills the forum requirements. I want to calculate the pairwise correlations across the (abnormal) returns of 80 firms,
for an event study test statistic and I need the average over all these pairwise correlations. I have read that I might achieve what I intend by using a loop, however I am unable to program
such a loop unfortunately. In the last clumn i have stored a value I also need to construct the test statistic, but for the correlations that value is irrelevant.

Below you can find an excerpt of my data. Any help would be highly appreciated!!
Jonas


input long Date float(resid_MarketModel_1 resid_MarketModel_2 resid_MarketModel_3 BMP_t)
22715 .028857823 -.012030476 .011276288 -17.066809
22718 .001680778 .015519869 .04662051 -17.066809
22719 -.013552702 -.0013050697 -.010423742 -17.066809
22720 .011290512 .0247944 .022345623 -17.066809
22721 -.017967667 .0013580762 -.02512785 -17.066809
22722 -.009187383 -.011011245 -.0087083 -17.066809
22725 .003578504 -.0034927686 -.00557095 -17.066809
22726 -.005407726 .010930496 .006511768 -17.066809
22727 -.013838122 -.01001027 -.015077838 -17.066809
22728 -.00786587 -.006517837 -.007176251 -17.066809
22729 .0239404 .003767889 .027022535 -17.066809
22732 -.011907137 -.014157334 -.02665999 -17.066809
22733 .005552202 -.008957997 -.009840348 -17.066809
22734 -.023531053 .00018480085 -.02001087 -17.066809
22735 .0040255897 -.017226135 -.009717827 -17.066809
22736 -.014861178 -.010953325 -.007508885 -17.066809
22739 -.012034866 -.003144226 -.01649329 -17.066809
22740 -.0007985968 -.001498356 .007198072 -17.066809
22741 -.0037047905 -.006126278 -.0025296016 -17.066809

[Jonas Petter]

In case someone is struggling with this too:

alpha resid_*, std det // average pairwise cross correlations
gen pwcorr_mean = r(rho)

does the trick.