Proc Nlmixed data=Mall tech=quanew method=gauss maxiter=1000;
parms B0=0, B3=0;
	p=1;
	   do i=0 TO t-1;
        logit=B0+B3*Robel;
	      p=p*(exp(logit)/(1+exp(logit)));
	   end;
model ifate~binomial(1,p);
RUN;