Dear All,    I want to impute the missing obs with previous value, but have no idea, though I referred below good link: https://communities.sas.com/t5/SAS-Programming/Last-Observation-Carried-Foward/m-p/309339#M66511   Yes, all the missing variable should be imputed from the previous value, as this would be the accumulative number of subjects till some timepoint.    The datafile in xlsx has been attached.    Thank you.      ... View more

Dear All, I have one very simple question:   Q1: How could below dataset be input successfully?  Here is the data I have, which is created by below code: data have; input subjid avisit dat $20.; datalines; 1001 0 1997-01-01 1001 1 1997-02-01 1001 2 1001 3 1997-05-01 1002 0 1998-01-01 1002 1 1998-02-01 1002 2 1002 3 1002 4 1002 5 1002 6 1998-12-01 1003 0 1999-01-01 1003 1 1999-02-01 1003 2 1003 3 1999-03-08 1003 4 1003 5 1999-04-05 1003 6 1999-05-01 1003 7 1003 8 ; run; data have_; set have; adt=input(dat, yymmdd10.); attrib adt format=yymmdd10.; run;     Q2: If you look at the last variable adt generated by below dataset WANT (which, unfortunately I can't produce by below code successfully, please assist me as well), you will see actually, I want to derive the last non-missing dat which should prior the consecutive dat, but I have no idea how to derive the last column, base on the first 3 variable correctly, hope you can help me.  data want; infile datalines delimiter=","; input subjid avisit dat $20. adt $20.; datalines; 1001,0,1997-01-01,1997-05-01 1001,1,1997-02-01,1997-05-01 1001,2, ,1997-05-01 1001,3,1997-05-01,1997-05-01 1002,0,1998-01-01,1998-02-01 1002,1,1998-02-01,1998-02-01 1002,2, ,1998-02-01 1002,3, ,1998-02-01 1002,4, ,1998-02-01 1002,5, ,1998-02-01 1002,6,1998-12-01,1998-02-01 1003,0,1999-01-01,1999-05-01 1003,1,1999-02-01,1999-05-01 1003,2, ,1999-05-01 1003,3,1999-03-08,1999-05-01 1003,4, ,1999-05-01 1003,5,1999-04-05,1999-05-01 1003,6,1999-05-01,1999-05-01 1003,7, ,1999-05-01 1003,8, ,1999-05-01 ; run;   ... View more

%let p0=0.05; %let p1=0.25; %let alpha=0.1; %let beta=0.1; %let usern=15; ****Stage 1; data stage1; do n1=5 to &usern-1;/*The totals ACCEPTABLE size is set sas "usern", n1 is the total size to be studied at stage 1*/ do r1=0; term1_p0 = cdf('BINOMIAL', r1, &p0, n1);/*PET: probability of terminate after stage 1 when the drug is ineffective;This is also the Type I ERROR at stage 1*/ term1_p1 = cdf('BINOMIAL', r1, &p1, n1);/*Probability of early ternation after stage 1 when, in fact, the drug is effective;*/ if term1_p1=<&beta then output; /*remove solution sets that do not meet the beta requirement; This is the type II ERROR occurred at stage 1*/ end; end; run; *****Stage 2; data stage12; set stage1; do n2=n1+1 to &usern-1; do r2=r1+1 to n2 while (r1<r2<n2); **term2_p0=0; *****initialize the summation terms for alpha & beta calculations; **term2_p1=0; do x1=r1+1 to min(r2, n1);/*x1: stand for the possible number of responses at stage 2 ONLY*/ **dum0=pdf('BINOMIAL', x1, &p0, n1)*cdf('BINOMIAL', r2-x1, &p0, n2-n1); **dum1=pdf('BINOMIAL', x1, &p1, n1)*cdf('BINOMIAL', r2-x1, &p1, n2-n1); **term2_p0= term2_p0+dum0;*recursive formulae used for summation terms; **term2_p1= term2_p1+dum1; output; end; output;/*Remove the conditions "1-(term1_p0+term2_p0)=<&alpha and" */ end; end; run; Regard to the above code, how could the value of X1 be greater than r2 as shown in below shot? ... View more