You can  run a join on the output of PROC SUMMARY ... View more

thank you so much   regards kajal ... View more

Your macro does not have any %DO loop.  So it cannot generate separate queries for each date value.   If your list of date values is small you could skip the macro and just put the code into the macro variable.  (has to be small since there is 64K byte limit on the length of a macro variable and the code is larger than just the date value).  Then just expand the macro variable to run the code. proc sql noprint; select distinct catx('create table',cats('sample_',date) ,'as select * from cars where date=',date,';') into :to_dt separated by ' ' from test ; &to_dt quit;   ... View more

@kajal_30 wrote: Exactly because I am trying a workaround and as I already clarified that I have values which I need to use to concatenate with other value to create a table name just a simple 1 line ask   have a value -2 associated to another value 1234  so I need to use 1234 in the table name that's it.   So why do you need a macro variable named -2? Why not create a lookup table where you look up -2 and then you retrieve the value 1234? All this can be done in a SAS data set, no macros needed.   You are too tightly focused on the need for a macro variable, which as we explained cannot be named -2. We are trying to propose other ways to get you there, but to do that, we need you to step back and explain the problem at a high level, and not at the "I must have a macro variable" level. This focus on a particular piece of coding you want is usually a bad thing, because it prevents us from suggesting better ways to handle the problem. This is called the XY problem and is not a good way to proceed. ... View more

@kajal_30 wrote: it is going be count(IDNUM) from master Please answer my other question: "Also, if there are 100 or more data sets, what is the data set name for the 100th data set?" ... View more

Why not just leave them as DATE values and attach format to print the values in that YYYYMM style? data two; set want; new_dt = active_dt ; old_dt = intnx('month',active_dt,-1); format new_dt old_dt yymmn6.; run; If you really need to create those integers in the YYY,YMM style then remember that the INTNX() interval MONTH works with DATE values.  Since you seem to have a date valued variable (otherwise the DATE9 format would print the values as gibberish) just use that variable with your INTNX() call. data two; set want; new_dt = input(put(active_dt,yymmn6.),6.); old_dt = input(put(intnx('month',active_dt,-1),yymmn6.),6.); run; If you only have the YYY,YMM integers then convert them into date values first. Example: 185 data test; 186 new_dt=202403; 187 date_value=input(put(new_dt,z6.),yymmn6.); 188 format date_value yymmdd10.; 189 put (_all_) (=); 190 run; new_dt=202403 date_value=2024-03-01   ... View more

@kajal_30 wrote: Hi @PeterClemmensen    As I said in the last line need abc_2408 which is the latest dataset having prefix abc That is an example, not a rule. Computer programs need rules to apply in more generic cases. For example, how do I know you wanted "abc_2408" if there was a "pdq_2408"? I think you need at least 2 rules, one involving the part before the _ (if any) and if 2408 is supposed to contain some sort of date information tell us how to parse it to determine "latest". ... View more

data f.rprt; set out.heads_20 out2.heads_21 out3.heads_23 out4.heads_26 out5.heads_28 indsname=indsname; by ty_number ve_date; orig_dsn=indsname; run;   ... View more

Thank you so much it also works    ... View more

data test; infile datalines ; input ID (status1-status4)($); datalines ; 1 yes no no yes 2 no yes no yes 3 yes yes no no 4 no no no no 5 no no yes no ; run; data allno; set test; if sum(status1='no',status2='no',status3='no',status4='no')=4; run; ... View more

@kajal_30 wrote: Thank you actually I overestimated myself. I thought I would be able to handle it but couldn't . Actually there are more than 1 column along with name so not sure how to handle multiple colums. Eg: dataset 1 has 10 columns and dataset 2 has 12 columns.   Not sure how to include rest of the columns in the same code you provided.   Regards Kajal What you can do with one variable you can do with other variables BUT naming will be different. So Transpose each of the additional variables in dataset2 into its own data set and merge all of them on the name value if you think you really need a data set.   The Report approach @Tom provided would with with multiple variables.   data dataset1; input ID name $; cards; 1 peter 2 john 3 muse 4 peena 5 tina ; data dataset2; input ID Value $ x $; cards; 1 cbz1 abc 2 cbz2 pdq 2 cbz3 xyz 2 cbz4 abc 3 cbz5 pdq 3 cbz6 xyz 3 cbz7 abc 4 cbz8 pdq 4 cbz9 xyz 4 cbz10 abc 4 cbz11 pdq 5 cbz12 xyz ; data want; merge dataset1 dataset2; by id; if first.id then row=1; else row+1; run; proc report data=want; column id name value,row x,row dummy; define id / group; define name / group; define row / across ' '; define dummy / computed noprint; compute dummy; dummy=0; endcomp; run; ... View more

Thank you ... View more

More information. More information. More information.   How big is "big"? How many observations, how many variables, what is the observation size? How large is t4? What is the type and size of rn? How many distinct values do you have? Where are these datasets really stored? WORK (as your code suggests), permanent SAS library, external database?   Hash method: data t1; set t2; where name like 'PNB%' and date <= "7dec2002:00:00:00"dt ; if _n_ = 1 then do; declare hash t4 (dataset:"t4 (keep=rn)"); t4.definekey("rn"); t4.definedata("rn"); t4.definedone(); end; if t4.check(); run; ... View more

Thank you @Tom  ... View more