I have a large dataset of stocks and daily returns. Because it is too large, I dont want to any analysis on the whole dataset. I would like to choose a subsample to play around with first. Say, I would like to take 50 stocks and I want all observations for those 50 stocks.
How can I achieve this?
So just pick 50 random stocks in your data set and retrieve every observation of those stocks, correct?
If so, then I think the Cluster Statement in PROC SURVEYSELECT is the way to go
data stocks(drop=i);
do i=1 to 1e5;
stock=rand('integer', 1, 1000);
output;
end;
run;
proc surveyselect data=stocks method=srs n=50 seed=123 out=want;
cluster stock;
run;
So just pick 50 random stocks in your data set and retrieve every observation of those stocks, correct?
If so, then I think the Cluster Statement in PROC SURVEYSELECT is the way to go
data stocks(drop=i);
do i=1 to 1e5;
stock=rand('integer', 1, 1000);
output;
end;
run;
proc surveyselect data=stocks method=srs n=50 seed=123 out=want;
cluster stock;
run;
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.