How to keep estimates to do my test ?

pmorel3 — Wed, 29 Jan 2020 21:19:40 GMT

I am a beginner in SAS.

I do not know how to keep all of my regression estimates in order to do my PROC test below.

Ultimately, I would like to compare 2 by 2 all of my 10 models by testing each time whether the 2 intercepts are statistically the same or not. That's why I prefer to compute delta before.

But the trouble is I have 12 main comparisons of these models to do and I don't want to write manually each estimate to compute my delta for each comparison.

Can you help me to write a code in order to link automatically my PROC test and all the regression estimates I need to compute my delta ?

proc format;           
	picture stderrf (round)       
     	low-high=' 9.99)' (prefix='(')                                
            .=' '; 
run;

ods output ParameterEstimates (persist) = t;                        
                                                                    
proc reg data = modelization ;                                              
      M1: model rp=rm;
	  MFS1: model rp=rm rm_zdy rm_ztbl rm_ztms rm_zdfy;

      M2: model rp=rm rmrsq;
      MFS2: model rp=rm rm_zdy rm_ztbl rm_ztms rm_zdfy rmrsq;

	  MPM10: model rp=rm rm_pred_mean;
	  MPM11: model rp=rm rm_pred_mean rmrsq;

	  MCFG1: model rp=rm zdy ztbl ztms zdfy rm_zdy rm_ztbl rm_ztms rm_zdfy;
      MCFG2: model rp=rm zdy ztbl ztms zdfy rm_zdy rm_ztbl rm_ztms rm_zdfy rmrsq;

	  MPM20: model rp=pred_mean rm rm_pred_mean;
      MPM21: model rp=pred_mean rm rm_pred_mean rmrsq;
run;                                                                
                                                                    
ods output close;                                                   
                                                                    
proc print data=t;                                                  
run;

proc tabulate data=t noseps;      
  class model variable;                      
  var estimate stderr;     
  table variable=''*(estimate =' '*sum=' '                          
                     stderr=' '*sum=' '*F=stderrf.),                
         model=' '                                                  
          / box=[label="Parameter"] rts=15 row=float misstext=' '; 
run;

/*Example of only 1 comparison by computing delta: difference of intercepts to make a test of significativity */
data modelization;
set modelization;
intercept23 = rp + 0.27098*rm + 0.001889*zdy - 0.12634*ztbl - 0.23400*ztms - 0.38475*zdfy + 0.27095*rm_zdy + 6.88259*rm_ztbl + 3.89297*rm_ztms - 0.72923*rm_zdfy + 0.73268*rmrsq; 
intercept24 = rp + 0.10546*rm + 1.68536*pred_mean + 6.90634*rm_pred_mean + 0.26742*rmrsq;
delta12 = intercept23-intercept24;
run;

   proc univariate data=modelization;
	  var delta12;
	run;

Re: How to keep estimates to do my test ?

pmorel3 — Thu, 30 Jan 2020 16:53:12 GMT

Is PROC SCORE suitable here ?

topic Re: How to keep estimates to do my test ? in Statistical Procedures

How to keep estimates to do my test ?

Re: How to keep estimates to do my test ?