Hi leva, You could declare a hiter object and use it with the remove method of the hash object. Hope this helps, Vasile ... View more

Hi, Here it is an idea using SQL statements: proc sql; create table temp as select name, year-age as yb from person where age is not missing; create table temp2 as select distinct name, yb from temp; update person set age=year-(select yb from temp2 where name=person.name ) where age is missing ; quit; Warm regards, Vasile ... View more

Hi, You provided a part of the desired outcome as: ......................... C 1999 37 C 2000 38 C 2002 39 Is that correct? Warm regards, Vasile ... View more

Or with a SQL statement: proc sql; select *, sum(hhincome) as income, sum(hhtype) as type from temp group by houseid ; quit; Warm regards Vasile ... View more

Hi, Start by summarizing the variable of interest, like: proc sql; select sum(var_data) as year_data from many_data; quit; Warm regards, Vasile ... View more

Hi, I hope that I understood your particular requirements and this will help. -- data temp; input houseid subjectid hhincome hhtype; datalines; 1001 10011 10000 2 1001 10012 . . 1002 10021 . . 1002 10022 25000 3 ; run; proc sort data=temp out=temp2; by houseid descending hhincome; run; data temp3; set temp2; by houseid; retain hhincome2 0; retain hhtype2 0; if first.houseid=1 then do; hhincome2=hhincome; hhtype2=hhtype; end; output; run; Warm regards, Vasile ... View more

Hi Valérie, I tested your code (excluding those supplementary where conditions) on SAS datasets and it seems to work fine. I would say you'd better and faster try to ask a question to SAS technical support about that. Warm regards, Vasile ... View more

Hi, In the case you still have the problem unsolved, it will be good to explain the purpose of the code (you are saying something in the subject line) and provide a sample of the data. Also the log could be useful. Warm regards, Vasile ... View more

Hi, You could use parameters and lags too: %macro lagi(nlags=2); %local i; data sumlags(keep= sum); set test ; sum=col1; %do i=1 %to &nlags; x=lag&i(col1); sum=sum+x; %end; if sum ne . then output; run; %mend lagi; %lagi(nlags=5); Warm regards, Vasile ... View more

Hi, About the storage and retrieving user-defined format you could read from: http://support.sas.com/kb/37/332.html It will be a good practice to create some documentation about the user-defined format and then you could share with all those using that format. Warm regards, Vasile ... View more

Hi, %let n=4; data selected; input Date Subject var; cards; 17837 1 1000 17867 1 1100 17898 1 1050 17929 1 1200 17957 1 1200 17988 1 1250 18018 1 1150 17837 2 1000 17867 2 1100 17898 2 1050 17929 2 1200 17957 2 1200 17988 2 1250 1 3 1200 18018 2 1150 ; proc sort data=selected; by subject date; data selected; set selected; by subject date; if first.subject then i=0; i+1; if i le &n then output; drop i; run; A more complex example could be found at: http://support.sas.com/kb/33/009.html Warm regards, Vasile ... View more

Hi, You could read about conditionally writing observations in more than one dataset at: http://support.sas.com/documentation/cdl/en/basess/58133/HTML/default/a001343763.htm Warm regards, Vasile data database; input month $3. pay bonus; cards; jan 20000 1000 jan 10000 1100 jan 5000 2000 feb 4000 3000 feb 5000 4000 mar 4800 600 mar 5000 800 ; data jan feb mar; set database; select (month); when ('jan') output jan; when ('feb') output feb; when ('mar') output mar; otherwise; end; run; ... View more

Hi Wendy, From that last SQL statement, do you think that by transposing the DATENAMES table you'd get a more useful result? Warm regards, Vasile ... View more

Hi, I am not yet an expert on SAS. Try look on proc format. Warm regards, Vasile ... View more