Hi Everyone, My data below has id, sign (+1/-1) and value and already in order of time.  The sign supposed to be in the +1 -1 +1 -1 sequence. However, there are rows having the same sign (+1 +1 +1). Thus I need to keep only 1 row to restore the right sequence rule.  So for +1+1+1..+1, I only keep the row with the highest value. This one below, I want to keep  2 3 1 8   2 1 1 0 2 2 1 5 2 3 1 8 And for -1-1-1...-1, I want to keep the row with the lowest value. This one below, I want to keep  1 5 -1 5 1 4 -1 8 1 5 -1 5 1 6 -1 7 Can you help me with this problem? Many thanks, HHC   data have; input id time sign value; datalines; 1 1 1 9 1 2 -1 -8 1 3 1 4 1 4 -1 8 1 5 -1 5 1 6 -1 7 2 1 1 0 2 2 1 5 2 3 1 8 2 4 -1 5 2 5 1 4 2 6 -1 3 2 7 1 5 ;run;     ... View more

Hi Everyone, I am just curious which method is fast: 1 - Put multiple (If else then) (If else then)  in 1 data step 2 - Using multiple data step and each contains only 1 (If else then)    Or in general when combining multiple operations in 1 data step is faster than breaking into multiple data steps. Thank you for your advice. HHC   ... View more

Hi, I run the below code and SAS return note like: Likelihood Optimization Algorithm Converged in 2 Iterations. NOTE: The above message was for the following BY group: ... 5020 options notes ; Is there any way for SAS not to print these lines out? I use "option nonotes" the the 2nd lines are not print, but the first line still there. Thank you, HHC proc ucm data=have2 noprint; by tic; id n interval=obs; model value; irregular; level variance=0 noest; slope plot=smooth; forecast lead=0 outfor=for plot=decomp; run;   ... View more

Hi, My code below helps to mark the top 3 value for each ID. However, as you see, SAS will assign top number for every rows which is unnecessary. For the sake of curiosity and efficiency, is there any way to tell SAS stop assigning value after top=3? Thank you, HHC   data have; input ID value ; cards; 1 5 1 8 1 2 1 0 2 6 2 10 2 9 2 1 2 5 2 33 run; proc sort data=have; by id descending value; run; data want; set have; by id; if first.id then top=0; top+1; retain top; run; ... View more

Hi Everyone, My data has date, time, ID and value. For each row, say date=1, time =12 and ID = 1, I want to find the 1st highest, 2nd highest and 3rd highest values for that id, of that date, and up to that time (<12). The illustration of that row at the end. Can you please help me? Thank you, HHC data have; input date time ID value; cards; 0 1 1 88888 0 1 9 585 1 1 1 999 1 2 1 7 1 3 1 3 1 4 1 101 1 7 1 102 1 9 1 12 1 12 1 9999 1 13 1 105 1 14 1 1 2 12 1 5 2 13 1 8 1 2 124 1 2 11 1 2 9999 1 2 9 ; data temp; set have; if id=1 and date=1 and time<12 ;run; proc sort data=temp; by id descending value;run; *pick to 3; data temp; set temp; if _N_<=3; run; *spread horizontally; proc transpose data=temp out=want (drop = _NAME_); by date id; var value;run; data want; set want; rename col1=top1 col2=top2 col3=top3; time=12;run;   ... View more

Hi Everyone, My data has date, ID and value. For each date and ID, say date=7 and ID = 1, I look back 5 days (from date = 6 to date = 2) and find the 1st highest, 2nd highest and 3rd highest values. So the wanted data should have 3 more new columns for these highest. Can you please help me? Thank you, HHC   data have; infile cards; input date ID value; cards; 1 1 1 2 1 7 3 1 3 4 1 101 5 1 102 6 1 103 7 1 104 8 1 105 8 2 1 9 2 124 10 2 11 11 2 9999 12 2 9 ; run; a  ... View more