I have a large dataset that contains daily returns for a many stocks. The dataset has 3 columns: date, stocks, and returns. I would like to calculate the rolling correlation and covariance between stocks for the last 1 week, 4 weeks (month), and 50 weeks (year). FYI, I am implementing a Corsi HAR model as I try to predict correlation next period using the correlation between stocks during the last 1 week, 4 weeks and 50 weeks. What would be the best approach? Thank you very much for your help! ... View more

I have the following input:   Ticker1 Ticker2 Edge ABC EFG 3 ABC NOM 10 ABC XYZ 50 EFG NOM 2 EFG XYZ 0 EFG QQQ 2 XYZ QQQ 6 XYZ NOM 99   Which I would like to convert to an n x n matrix:       ABC EFG NOM XYZ QQQ ABC   3 10 50   EFG 3   2 0 2 NOM 10 2   99 6 XYZ 50 0 99     QQQ   2 6       How should I do this? the above is just to give an idea. My input dataset is:  data WORK.CLASS; infile datalines dsd truncover; input ticker1:$10. ticker2:$10. edge:32.; datalines4; ABC,EFG,3 ABC,NOM,10 ABC,XYZ,50 EFG,NOM,2 EFG,XYZ,0 EFG,QQQ,2 XYZ,QQQ,6 XYZ,NOM,99 ;;;; I cut this down to 15 obs but my dataset has 500 and the ticker set are the same for ticker1 and ticker2. Thus I would want a n x n matrix.    ... View more