Recently, I am studying the non-inferiority test based on the difference of two proportions. The teacher asked us to use Chi-square test do, but I used SAS, and found it use the Wald test do this by default, and can't set the method as Chi-square, hence I want to bring this out to seek: Is Wald Test a kind of Chi-square test?
The code is as below:
data pd1;
input grp $ response count;
datalines;
grp1 1 20
grp1 0 4
plcb 1 18
plcb 0 6
;
run;
/*Do non-inferiority test by Chi-square test*/
proc freq data=pd1;
tables grp*response/chisq alpha=0.025 riskdiff(column=2 noninf correct margin=0.3);/*noninf:non-inferiority test; margin: the margin want to detect*/
weight count;/*count: then number of responders*/
run;