I was trying to create some macro to select the first non missing value for every variable of interest per subject.   After sorting the data... %fstnmissvalue(dsn,var); data temp..&var ; set &dsn (keep=bene_id A0310F_ENTRY_DSCHRG_CD TRGT_DT rename=(TRGT_DT=&VAR._DT A0310F_ENTRY_DSCHRG_CD=&VAR._ENTRY_DSCHRG_CD)) ; where &var ne . and &var ne ""; by bene_id; if first.bene_id; run; %mend; %macro fstnmissvalue_loop(varlist); %let n1=%sysfunc(countw(&varlist)); %let dsn1=temp.mds_asmt11_sum; %do i=1 %to &n1; %let var1=%scan(&varlist, &i); %fstnmissvalue(&dsn1,&var1); %end; %mend; %fstnmissvalue_loop(G0110H1_EATG_SELF_CD G0110B1_TRNSFR_SELF_CD G0110I1_TOILTG_SELF_CD G0120A_BATHG_SELF_CD );   Notes from the log:   785 %fstnmissvalue(dsn,var); - 180 WARNING: Apparent invocation of macro FSTNMISSVALUE not resolved. ERROR 180-322: Statement is not valid or it is used out of proper order. 786 data temp..&var ; ----- 22 201 WARNING: Apparent symbolic reference VAR not resolved. 787 set &dsn (keep=bene_id A0310F_ENTRY_DSCHRG_CD TRGT_DT rename=(TRGT_DT=&VAR._DT - ---------------------- 214 79 23 ERROR 22-322: Syntax error, expecting one of the following: a name, a quoted string, /, ;, _DATA_, _LAST_, _NULL_. ERROR 201-322: The option is not recognized and will be ignored. ERROR 214-322: Variable name & is not valid. ERROR 79-322: Expecting a =. ERROR 23-7: Invalid value for the RENAME option. ------- 23 787! set &dsn (keep=bene_id A0310F_ENTRY_DSCHRG_CD TRGT_DT rename=(TRGT_DT=&VAR._DT ------- 214 787! A0310F_ENTRY_DSCHRG_CD=&VAR._ENTRY_DSCHRG_CD)) ; WARNING: Apparent symbolic reference VAR not resolved. WARNING: Apparent symbolic reference VAR not resolved. ERROR 214-322: Variable name VAR._DT is not valid. 788 where &var ne . and &var ne ""; WARNING: Apparent symbolic reference VAR not resolved. 788 where &var ne . and &var ne ""; - 180 ERROR 180-322: Statement is not valid or it is used out of proper order. WARNING: Apparent symbolic reference VAR not resolved. WARNING: Apparent symbolic reference VAR not resolved. ERROR: Syntax error while parsing WHERE clause. 789 by bene_id; 790 if first.bene_id; 791 run; NOTE: The SAS System stopped processing this step because of errors. WARNING: The data set WORK.VAR may be incomplete. When this step was stopped there were 0 observations and 0 variables. WARNING: Data set WORK.VAR was not replaced because this step was stopped. NOTE: DATA statement used (Total process time): real time 0.09 seconds cpu time 0.03 seconds ERROR: No matching %MACRO statement for this %MEND statement. 792 %mend; 793 794 795 %macro fstnmissvalue_loop(varlist); 796 %let n1=%sysfunc(countw(&varlist)); 797 %let dsn1=temp.mds_asmt11_sum; 798 %do i=1 %to &n1; 799 %let var1=%scan(&varlist, &i); 800 %fstnmissvalue(&dsn1,&var1); 801 %end; 802 %mend; 803 804 /*%let dsn=temp.mds_asmt11_sum;*/ 805 %fstnmissvalue_loop(G0110H1_EATG_SELF_CD G0110B1_TRNSFR_SELF_CD G0110I1_TOILTG_SELF_CD G0120A_BATHG_SELF_CD ); NOTE: Line generated by the invoked macro "FSTNMISSVALUE_LOOP". 1 %fstnmissvalue(&dsn1,&var1); - 180 WARNING: Apparent invocation of macro FSTNMISSVALUE not resolved. ERROR 180-322: Statement is not valid or it is used out of proper order. WARNING: Apparent invocation of macro FSTNMISSVALUE not resolved. NOTE: Line generated by the invoked macro "FSTNMISSVALUE_LOOP". 3 %fstnmissvalue(&dsn1,&var1); - 180 ERROR 180-322: Statement is not valid or it is used out of proper order. WARNING: Apparent invocation of macro FSTNMISSVALUE not resolved. NOTE: Line generated by the invoked macro "FSTNMISSVALUE_LOOP". 5 %fstnmissvalue(&dsn1,&var1); - 180 ERROR 180-322: Statement is not valid or it is used out of proper order. WARNING: Apparent invocation of macro FSTNMISSVALUE not resolved. NOTE: Line generated by the invoked macro "FSTNMISSVALUE_LOOP". 7 %fstnmissvalue(&dsn1,&var1); - 180 ERROR 180-322: Statement is not valid or it is used out of proper order. I have no idea where went wrong with my code. Is there any one that could help with my problem? I'll be very appreicated about your help. Thanks!!!   ... View more

I was trying to minic the macro code (link as below) to create a table, which is supposed to display median, qrange as well as p value from wilcoxon test for a continous variable called 'sum_PT_mins_1yr'  by a categorical variable called 'race'.   http://www.lexjansen.com/nesug/nesug08/ff/ff06.pdf   My code is below: %macro continuous(cpred, i); proc univariate data=temp3.PTestimate (where=(race in (1, 2) and &cpred>0))) noprint; class race; var &cpred; output out=Quantilestatistics&i median=median qrange=qrange; run; proc npar1way data=temp3.PTestimate (where=(race in (1, 2) and &cpred>0))) wilcoxon; class race; var &cpred; output out=wilcoxon&i(keep=P2_WIL) wilcoxon ; run; data pvalue&i; set wilcoxon&i; if 0.0001<=P2_WIL< = 0.05 then p_value = (round(P2_WIL,0.0001)||'*'); else if P2_WIL<0.0001 then p_value = ('<0.0001'||'*'); else p_value = round(P2_WIL,0.0001); keep p_value; run; data descript&i; merge Quantilestatistics&i pvalue&i; cmedian = put(median,8.2); cqrange = put(qrange,8.2); median_qrange = (trim(left(cmedian))) || ' (' || trim(left(cqrange)) || ')'; drop median qrange; run; proc transpose data=descript&i out=tdescript&i; var median_qrange; copy p_value; run; %mend continuous; %macro contiunous (sum_PT_mins_1yr, 1); However, I got an error message from the log:   ERROR: Invalid macro parameter name 1. It should be a valid SAS identifier no longer than 32 characters. ERROR: A dummy macro will be compiled.   I have no clue how to fix this problem. Probably I have more errors than that in this code... Your suggestions would be highly appreicated. Thank you. ... View more

I'm struggling with the methodology and technique metioned below. If there is any one who has some ideas on how this work with SAS Macro or kindly point me to the right direction to google the 'key words/terminology', that would be a great help! Thanks!   The purpose of doing it is to show the contribution to the predictiveness of adding the 6 selected variables.   1. First, generate 500 bootstrap samples of both the new model B and new model C and “freeze the samples.”  2. For each bootstrap sample you compare the model B and model C chi-square.  There will be 500 pairs of chi square differences, which represents the population of chi-square differences.   3. We can then compare with a chi-square test using the number of degrees of freedom of the selected variables included.  So, for example, if 6 variables are included in the final model after bootstrapping, we would run a chi square test with 6 degrees of freedom.    You valuable inputs would be highly appreciated! Thank you.   ... View more