Hi Sam, Please try the below code data have;     infile datalines missover;     input pid  reason $ eventdt : date9. samdt : date9.;     format eventdt samdt date9.; datalines; 100  xxxy    23aug2011  14aug2011 100  xxxy    19dec2011 110  yyyy    31jul2010 120  zzzzz   14jun2010 120  zzzzz   27oct2010  28oct2010 120  zzzzz   02jan2011  17feb2011 ; run; proc sort data=have;     by pid; run; data want(keep=pid reason samdt flag);     set have(where=(samdt ne .));     retain new;     by pid reason;     flag=1;     if first.reason then output want; run; data want_;       merge have(in=a) want(in=b rename=(samdt=new_date));     by pid reason;     if a;     if flag=. then flag=0; run;                           Obs pid reason  eventdt             samdt         new_date    flag                            1 100 xxxy  23AUG2011 14AUG2011 14AUG2011  1                            2 100 xxxy  19DEC2011        . 14AUG2011  1                            3 110 yyyy  31JUL2010        .        .  0                            4 120 zzzzz 14JUN2010        . 28OCT2010  1                            5 120 zzzzz 27OCT2010 28OCT2010 28OCT2010  1                            6 120 zzzzz 02JAN2011 17FEB2011 28OCT2010  1 Thanks, Jagadish ... View more

Hi Sandhyaa, I am not sure if it is possible to get the seq no as you want in DI studio without coding. May be other members can you help you if they have come across similar situation. Here is a small piece of code which i hope can help you data have;      input src_col $; datalines; a a a b b c ; proc sort data=want;     by src_col; run; data want;     set have;     retain seq_no;     by src_col;     if first.src_col then seq_no=1;     else seq_no+1; run; Thanks, Jagadish ... View more

Hi Namrata, to avoid that error note as suggest by few of the community members you need to sort the wb1 data on country and newyear.  This will avoid the error. However there is another point which i would to inform you is that, in your post you mentioned that when you tried to import the csv file an error popped up that the year was both numeric and character. It some time happens that in a particular column in excel or csv file we find that there are numeric and character values. While importing sas will check the first 8 rows and will consider the data type based on that, for example if we have first 8 records with numeric and next character, sas will take the remaining records are numeric and display them as missing as character values are there. to avoid this we have a statement in proc import, which is MIXED=YES; if you use this you can avoid that message and there will be no need for you create another variable. Please try it and let me know your feedback Good Luck Jagadish ... View more

Hi , Please include the below statement in the proc import code, this will convert the numeric data to character values. This will help you to change the variable into character where there is numeric as well as character data, other columns will be as they are. MIXED=YES; Thanks, Jagadish ... View more

Hi Karthik, did you try exporting the file to dbms=xlsx Thanks, Jagadish ... View more

Hi Karthik, The code what you have written is perfectly fine, the problem occurs when you try to open the excel in excel 2007 and excel 2010. If you try to open the same file with excel 2003 it will open without any problem. May i know which is the excel version you are using (2007 or 2010). To avoid the problem add version=2003 to the code. PROC EXPORT DATA=last OUTFILE='/home/data.xls' DBMS=XLS REPLACE; SHEET="data"; version=2003; RUN; Hope this works for you. Good Luck Thanks, Jagadish ... View more