I actually could distinct select all the columns of interest but the thing is that since there are many of them I thought if its possible to make a shortcut by selecting the entire row corresponding to the name/date combination... ... View more

Hi Patrick and thanks for the reply. I did the select distinct but it only gave me the 2 selected columns, but what I would like to get if possible is all the columns (like in the small example that I put in the question), in other words to delete all the name/date combination duplicates I also tried to do data board_summary6; set board_summary6;   by comp_name date; if first.comp_name and first.date; run; but got an error message ... Thank you ... View more

Thanks!! ... View more

Hi Arthur, I actually figured out how to do my second little extra question, its quite simple if I'm correct and is of the following: data a; input name$ nu ; datalines; a 1 s 3 d . g 4 t . kk . f 6 ; run; data b; set a; if nu ne " "; run; thanks! ... View more

Hi arthur, I have millions of lines of code, and in a special column of interest there are around 1000 rows with nonmissing data, and the rest are missing. Is it possible to know where those nonmissing values are, i.e to output their observation number? Thank you ... View more

Hi Arthur and Patrick, thanks for the quick reply and both of your codes are very helpful!! Just some small extra questions if I may: when doing the proc freq, I realized that some columns which I thought were completely empty and which I was going to delete actually had some data, albeit small, which I might use in the future. 1) Is it possible to see what is the number of the observation where these data are present ? 2) is it possible to make an output file of all the columns but conditional on data being present on columns of my choice? Again thank you very much!!! ... View more

Thank you Hai.kuo and stat@sas !!! ... View more

Hi stat@sas, I did the following including your code: data dup; input id name$; datalines; 3 a 3 a 1 d 5 e 5 e 4 y 2 t 2 t ; run; data dup2; set dup; by id; if first.id; run; But I get an error message: ERROR 180-322: BY variables not properly sorted on dataset DUP And the result that I get is: id name 1 3 a It seems that the duplicate was deleted for the first row, but after that the code stopped functioning Thank you ... View more

thanks Jagadishkatam, you taught me a new function! ... View more

Thanks a lot stat@sas! ... View more

Hi Reeza its me again, now I would like to add another level of complexity, if I may: assume that in the previous case I had to calculate the cumulative sum for one company, but now I need to do the same thing but for many companies, so each time I get to a new company the cumulative sum restarts calculating. Thank you ... View more