Save your code and data. then restart SAS, or disconnect and reconnect SAS Studio. At some point, it looks like you have an unmatched single or double quote.   From what you posted, it also looks to me like you called a macro named logistic_aic_sbc_score. The unbalanced quote may be in that macro..   SteveDenham ... View more

@SIMMII wrote: The log is generating hundreds of other errors, if I include a class statement do you think it will fix most of my other errors? If so, where should I include the class statement in the macro? The CLASS statement should fix the X5 problem. With regards to the other errors, it is impossible for me to answer since you didn't show us the other errors.   The CLASS statement should go in the macro in the exact same place as it goes in your non-macro code. You probably need modify the macro definition to include a macro parameter containing the list of CLASS variables as well.   When you run a macro in SAS, you ought to turn on macro debugging tools, specifically run this line of code before you run the macro   options mprint;       Although I question why you need a macro here when you already have written PROC LOGISTIC code that seems to do what you want. ... View more

I see, thank you very much! I'll use the insert code feature next time, sorry about that! ... View more

Something along the line of below sample code should do data work.ps2renamed; x41=1; y='Yes'; output; Y='No'; output; run; proc format; invalue yn_char2num(upcase) 'Y','YES' = 1 'N','NO' = 0 other =. ; run; data work.ps2renamed; set work.ps2renamed(rename=(y=y_char)); y=input(y_char,yn_char2num.); drop y_char; run; proc print data=work.ps2renamed; run;       ... View more

This works regardless of which regression procedure you use, it will work even if you don't use PROC REG. https://blogs.sas.com/content/iml/2014/02/17/the-missing-value-trick-for-scoring-a-regression-model.html   Side note: I am skeptical that a predictive model with 36 explanatory variables is a good thing to do, because of likely multi-collinearity between the predictor variables could make the model coefficients go crazy, and could inflate the variance of the predicted values. Similarly, if you have 19 explanatory variables, you are quite likely in the same situation. ... View more

If you just want the MEAN statistic then either subset the current summary dataset you have generated.  You could add a WHERE statement to your PROC TRANSPOSE step. where _stat_='MEAN'; Or you could just generate only the MEAN statistic into the summary dataset by changing the OUTPUT statement. proc means .... ; var x1-x34 ; output out=mean_summary mean=; run; That will make a dataset with only one observation with variables names X1-X34 that contain the MEAN value of the respective original variables.   Try it yourself: proc summary data=sashelp.class; var age height weight; output out=stat1; output out=stat2 mean=; run; proc print data=stat1; run; proc print data=stat2; run; proc transpose data=stat1 out=tran1; var age height weight; where _stat_='MEAN'; run; proc transpose data=stat2 out=tran2; var age height weight; run; proc print data=tran1; run; proc print data=tran2; run; TRAN1 and TRAN2 have the same results. ... View more