Dear Statalist forum,
I have quarterly data (Year_Quarter) from 1965 until 2020 and I want to generate a four period aggregated return. I first built decile portfolios based on characterstics and then generate the average return for each portfolio by the deciles. But if I want to generate the aggregated portfolio return over four quarters I get a lot of missing values, probably because most portfolio returns are not available in four following quarters. How can I solve the problem in a way that I still get a value of four period returns? Maybe if one, two or three returns are missing ad the average of the existing returns?
I used the following code for the Asset Growth portfolio:
bys AG10: gen agreg_av_return_AG= av_return_AG + av_return_AG[_n-1]+ av_return_AG[_n-2] + av_return_AG[_n-3]
I have also tried rangestat but it seems to produce completly wrong results:
rangestat (sum) agreg_av_return_AG= av_return_AG, by(AG10) int(Year_Quarter -4 0)
I would appreciate any help. Thanks in advance!
Related Posts with Four period aggregated return
Encoding strings stored as negative numbersHello I noticed something which puzzled me when trying to convert a variable stored partly as negat…
Rolling average using a variable as weight with tssmooth or otherI would like to calculate the rolling average of the estimated beta coefficient via statsby (say "_b…
Overlapping maps from two different coordinate datasets.I have constructed two maps of Nepal both attached below. One reflects 7 provinces while the other r…
test posting results Code: Cohort | Freq. Percent Cum. ------------+---------------------------------…
Bug in "tsegen"There is a bug in "tsegen" when the number of lags increase above 100. Is there any way to fix it? …
Subscribe to:
Post Comments (Atom)
0 Response to Four period aggregated return
Post a Comment