Dear Mkeintz Thank you very much for your detailed explanations! Only got back to SAS community after an extended span of research. Wish you a very good rest of the day! KS -, ... View more

Dear mkeintz,    Thank you for helping me a second time!  Your codes work perfectly.    The codes you wrote are very short and exquisite.  Do you mind if I ask you a further question?  Today I studied hash object for the first time, and looked at your codes and comments again.  But what I don't understand from your comments are:  The h.find() method returns a zero for success and a non-zero for failures.  This will retrieve the first instance of each CUSIP. The "until (h.find.next)^=0" performs a h.find_next() method at the bottom of each loop iteration, and returns a zero for successful retrieved of the next item of the same cusip.  How does h.find() function work exactly?  It seems to be used in a variety of merging, left full join, etc.  You can give me a quick answer, cos I am ashamed to take your precious time;)    Many thanks!  KS -    ... View more

Thank you Reinhard! I will apply them for my future analysis! Sincerely, KS -, ... View more

Hi, yabwon! Thank you very much, yabwon! I will download the package and carefully study them. I really appreciate your help and discussions! Sincerely, KS -, ... View more

Thank you mkeintz! I change that part of codes to this: (drop=v rc _tmps:). Now it cleanly works! Have a good day ! Sincerely, KS -, ... View more

Dear mkeintz I have an additional question. I ran your codes, and it worked well. Except one thing: I discovered that in my end-results all variables starting with _(number) disappeared! Maybe should I change the variable names? Or, is there is cure for this? Many thanks, KS Choi -, ... View more

Dear mkeintz,    Now it works perfectly!  Thank you so much , and Happy New Year!    Sincerely,  KS -,    ... View more

Thank you for your elaborate codes!    But my variables are over 4000, and they all have different names.  I will keep your codes for the future reference, though   KS -,  ... View more

To answer your quick question: The names of the variables look like these: _00036020 Num 8 BEST. _00036110 Num 8 BEST. _00037520 Num 8 BEST. _00108410 Num 8 BEST. _00154710 Num 8 BEST. _00212110 Num 8 BEST. SSBSQXJ0 Num 8 BEST. SSBXBZB0 Num 8 BEST. SSBYL6RK Num 8 BEST. SSBYPC1T Num 8 BEST. SSBZ0Y0T Num 8 BEST. Y2106R11 Num 8 BEST. Y2573F10 Num 8 BEST. ....... They are all numeric variables. ... View more

Dear mkeintz,    Oh, thank you for pointing out what I am missing!  So, today, I struggled to attach a prefix "Var_" to all the variables, using a set of codes provided by SAS support.     /* Rename all the variables, adding prefix "Var_" */ proc sql noprint; select cats(name,'=','Var_',name) into :list separated by ' ' from dictionary.columns where libname = 'WORK' and memname = 'Matrix_2016'; quit; proc datasets library = work nolist; modify Matrix_2016; rename &list; run; quit;    However, I ended up with another error message like this:    156 proc datasets library = work nolist; 157 modify Matrix_2016; WARNING: Apparent symbolic reference LIST not resolved. 158 rename &list; run;  I am sorry to take your precious time, but I really appreciate your help!    Sincerely,  KS -,    ... View more

Dear mkeintz,    Thank you so much for your help! I really appreciate it.    I ran the codes you provided, but there seems to be a small obstacle.  I got the following warning (bold) in the first part of your codes,    119 data _null_; 120 set Matrix_2016 nobs=n_analys; 121 call symput('n_analysts',cats(n_analys)); 122 array cusip_ids {*} var: ; WARNING: Defining an array with zero elements. 123 call symput('n_cusips',cats(dim(cusip_ids))); 124 stop; 125 run; NOTE: There were 1 observations read from the data set WORK.MATRIX_2016. NOTE: DATA statement used (Total process time): real time 0.01 seconds cpu time 0.01 seconds   And running the latter part of your codes, I receives the following error message.  ERROR: Invalid dimension specification for array vars. The upper bound of an array dimension is smaller than its corresponding lower bound. ERROR: Too few variables defined for the dimension(s) specified for the array vars. ERROR: Invalid dimension specification for array _tmps. The upper bound of an array dimension is smaller than its corresponding lower bound.    Can you help me a little bit more with this? I would greatly appreciate it,   Thank you !   Sincerely,  KS -,    ... View more

Oh, sorry, maybe I am mistaken..    Why some values are filled, whereas others are missing, is simply by chance (from the database).     ANALYS 1-10 are analyst ID. Var1-Var3 are stock1, stock2, and stock3.  So, the dataset is reporting that stock1 is taken care of by analyst 5, stock2 by analysts 3,7,9, stock3 by analysts 7,9,10.  What I want to do is to drag down the values to the right places corresponding to ANALYS.    ANALYS var1 var2 var3       1         5      3       7       2          .      7       9       3          .      9      10       4          .      .        .       5          .      .         .       6          .      .         .       7          .      .         .       8          .      .         .       9          .      .         .       10        .      .         .   Am I answering you question correctly?    Thanks,  KS -,  ... View more

Thank you for your reply!    The story about this dataset is that I wanted to transform a panel data of forecasting analysts and their firms into a matrix with a row of analysts and a column of firm identifiers.  From the SAS window, Original looks like this: (It has 6,263 obs. and 4,589 variables)  After I disperse the values to their appropriate rows for all the variables, I am going to change the missing to 0, the non-missing to 1.    Many thanks!  KS -,    ... View more