@Jose_Roberto wrote: Bom dia sou iniciante no SAS e estou coma seguinte dificuldade - Transformar o trecho do código sql abaixo para o formato do proc sql no SAS. No código SQL ficaria assim: where cod_operadora || '|' || num_contrato in ('003 | 123456789', '001 | 987654321') Detalhe: o campo cod_operadora é um campo texto e o campo num_contrato é um campo numérico. Como eu poderia converter o trecho do sql para o proc sql do SAS, conseguem me ajudar?   Translated via Google from Portuguese to English--not sure how correct it is, but it translated it well enough for me to understand. Good morning I am a beginner in SAS and I am having the following difficulty - Transform the sql code snippet below to the format of the sql proc in SAS. In the SQL code it would look like this: where cod_operadora || '|' || num_contrato in ('003 | 123456789', '001 | 987654321') Detail: the cod_operadora field is a text field and the num_contract field is a numeric field. How could I convert the sql snippet to the SAS sql proc, can you help me?  Here's my solution. Try running this code. / Aqui está minha solução. Tente executar este código.   data have; input cod_operadora :$3. num_contrato; datalines; 003 123456789 001 987654321 002 929193921 006 999239123 ; run; proc print data = have; run; proc sql; select * from have where catx(" | ", cod_operadora, num_contrato) in ("003 | 123456789", "001 | 987654321"); quit; The CATX function trims leading and trailing blanks, and inserts the " | " delimiter. / A função CATX apara os espaços em branco à esquerda e à direita e insere o caractere " | " delimitador.   ... View more

Do you only want the number of observations? Do you want them labeled in any way? There's a lot of interpretation that we have to do from your question. The below example concatenates all the data sets in the SASHELP library and their respective observations into a macro variable called "observations": proc sql noprint; select catx("-", memname, nobs) into: observations separated by ", " from dictionary.tables where libname = "SASHELP" and memtype = "DATA"; quit; %put &observations; We need more information though. Just number of observations into separate macro variables would be the following: proc sql noprint; select nobs into: observations1 - from dictionary.tables where libname = "SASHELP" and memtype = "DATA"; quit; %put _user_; ... View more

Certainly. I'll update the code to include the >= operator. ... View more

This may need to be modified in light of the comments above, but this gives what you requested, but it looks a little ugly in my opinion. data have; input patient $ 1 @3 month monthcnt; datalines; A 1 1 A 2 2 A 3 3 A 4 4 A 5 5 A 6 6 B 1 1 B 3 2 B 4 3 B 5 4 B 6 5 C 7 1 C 8 2 C 9 3 D 1 1 D 4 2 D 5 3 D 6 4 D 7 5 D 8 6 ; data have_2; set have; by patient month; lag_month = lag(month); if first.patient then call missing(lag_month); if not first.patient then do; if month - lag_month ^= 1 then skip_month = 1; else skip_month = 0; end; run; proc summary data = have_2 noprint; class patient; /* create table with patient as the identifier */ var skip_month; /* create column based on whether the individual has a skip month */ ways 1; /* eliminate default summary row for all records */ output out = want (drop = _type_ where = (total_months >= 6 or (total_months = 5 and skip_month = 1)) rename = (_freq_ = total_months)) /* (1) drop _type_, (2) subset based on months and skip_month, (3) rename default column _freq_ to total_months just so it looks nicer */ max = ; /* output the max skip_month for an individual */ run; Obs patient total_months skip_month 1 A 6 0 2 B 5 1 3 D 6 1     ... View more

What do you want your data to look like? I'm getting a little lost between all the output statements and arrays. That code seems a bit convoluted.   This flags whether a month was skipped, but I couldn't provide an answer without knowing what you want it to ultimately look like.   data have; input patient $ 1 @3 month monthcnt; datalines; A 1 1 A 2 2 A 3 3 A 4 4 A 5 5 A 6 6 B 1 1 B 3 2 B 4 3 B 5 4 B 6 5 C 7 1 C 8 2 C 9 3 D 1 1 D 4 2 D 5 3 D 6 4 D 7 5 D 8 6 ; data have_2; set have; by patient month; lag_month = lag(month); if first.patient then call missing(lag_month); if not first.patient then do; if month - lag_month ^= 1 then skip_month = 1; else skip_month = 0; end; run; You could use PROC MEANS to get the max for each ID and then merge it back. That or PROC SQL. Nonetheless, please also provide a DATALINES statement with your 'want' data set.   ... View more

I don't often use DO loops, so I may be violating some principles. Results weren't as linear as I prefer, but it seems to match what you want. Earlier results may be more comprehensive or streamlined.   data have2; set have; by id; if last.id then do; do until (date = enddate); date = intnx("month", lag(date), 1); output; end; end; run; data want; set have have2 (where = (date ^= .)); proc sort; by id; run; ... View more

What variable are you trying to update? Is it the one with the (...) prefix? Please post an example using the DATALINES statement so we can see the exact issue. ... View more

If I understand correctly, this is what you want:   data have; input record a :yymmdd10.; format a yymmdd10.; datalines; 1 2003-05-21 ; run; data want; set have; want = put(a, yymmn6.); run; ... View more