Hello,  This is my first attempt PROC OPTMODEL and would appreciate any help to solve the problem at hand. Through optimization, I am try to determine the optimal cut-off for each bucket such that the error is minmized between avg. PD and the avg. Def. Below are the datasets I have   Dataset 1       Dataset 2 Customer  PD Def   Bucket 0001 0.24% 0   1 0002 0.02% 0   2 0003 0.89% 0   3 0004 3.50% 0   4 0005 6.80% 1   5 0006 0.008% 0     0007 15.20% 1     0008 28.20% 1     0009 2.23% 0     0010 1.25% 1       I want to determine the optimal PD value for each bucket such that 1. The error, difference between avg. PD and avg Def, is minimized for each bucket. 2. The sum of error across buckets is minimized.   Thanks for your help ... View more

Hello Everyone,   The Have dataset is sorted by ID and Ranks. For each group, by ID, I am trying to pick the first observation for which the difference is less than equal to 12. For example, for ID 0001, the Want dataset has the secon observation. Would really appreciate if anyone could help me out.   Thanks Have       Want     ID Rank Diff   ID Rank Diff 0001 2 20         0001 3 8   0001 3 8 0001 4 44         0001 5 38         0002 2 18         0002 4 3   0002 4 3 0002 5 6         0002 12 30         0003 1 11   0003 1 11 0003 2 23         0003 3 47         0003 5 2         ... View more

Hello Everyone,   I am trying to down sample a dataset from yearly frequency to monthly. I would really appreciate if some one could guide me to do so in a data step. I will explain what I am trying to do with an example:   Have                                                        ID Date Rank 0001 12/31/2012 2 0001 12/31/2013 2 0002 6/30/2006 4 0002 12/31/2007 5   Need ID Date Rank 0001 12/31/2012 2 0001 1/31/2013 2 0001 2/28/2013 2 0001 3/31/2013 2 0001 4/30/2013 2 0001 5/31/2013 2 0001 6/30/2013 2 0001 7/31/2013 2 0001 8/31/2013 2 0001 9/30/2013 2 0001 10/31/2013 2 0001 11/30/2013 2 0001 12/31/2013 3 0001 1/31/2014 3             0001 11/30/2014 3       Similarly for the other ID   Thanks ... View more

Hello,   I am trying to execute the below code but the execution stops unexpectedly. For reference, the min_date is 15Feb1993 and the max_date is 15Dec2013. At the bottom of the macro I have also pasted macro variable resolution from the log. Any help is greatly appreciated. Thanks   %macro run_all; proc sql noprint; select max(datepart(current_date)) format=date9., min(datepart(current_date)) format=date9. into :max_date, :min_date from edf; quit; %do %until ("&min_date."d ge "&max_date."d); %let var_2= %sysfunc(inputn(&min_date,date9.),monyy5.); %cohort(&min_date,&var_2); %let min_date=%sysfunc(intnx(month,"&min_date."d,3,same),date9.); %put &min_date.; %end; %mend run_all;   SYMBOLGEN: Macro variable MIN_DATE resolves to 15FEB1993 MLOGIC(RUN_ALL): %PUT &min_date. SYMBOLGEN: Macro variable MIN_DATE resolves to 15MAY1993 15MAY1993 SYMBOLGEN: Macro variable MIN_DATE resolves to 15MAY1993 SYMBOLGEN: Macro variable MAX_DATE resolves to 15DEC2013 MLOGIC(RUN_ALL): %DO %UNTIL() condition is TRUE; loop will not iterate again.   ... View more