e****t 发帖数: 766 | 1 SAS 9.2 gives a way for noninferioriy test for the difference of two
proprotions.
I try to use that, but confused where to specify the command.
Does anyone use it before ?
here is the link for introduction
http://support.sas.com/documentation/cdl/en/procstat/63104/HTML/default/viewer.htm#procstat_freq_sect010.htm#procstat.freq.binopts
i tried specify in "proc freq" "tables" "test" comond, none of them work.
can anybody like to share the code ?
Thanks in advance! |
s*r 发帖数: 2757 | 2 没用过,不过那一整个webpage说的都是tables后面的option
【在 e****t 的大作中提到】 : SAS 9.2 gives a way for noninferioriy test for the difference of two : proprotions. : I try to use that, but confused where to specify the command. : Does anyone use it before ? : here is the link for introduction : http://support.sas.com/documentation/cdl/en/procstat/63104/HTML/default/viewer.htm#procstat_freq_sect010.htm#procstat.freq.binopts : i tried specify in "proc freq" "tables" "test" comond, none of them work. : can anybody like to share the code ? : Thanks in advance!
|
P****D 发帖数: 11146 | 3 proc freq;
table xxx / binomial(noninf);
run; |
e****t 发帖数: 766 | 4 Thanks all. it works now!
I specify similiar in riskdiff (option) ;
proc freq data = a ;
table x * y / riskdiff (noninf margin = 0.05) ;
run; |
e****t 发帖数: 766 | 5 sorry to bother again,
in the output, i didn't find p value.
Only CI is given. (i know i can invert it to get SE...lazy to do it ) can we
get non inferiority p value from SAS directly?
proc freq data = a ;
table x * y / riskdiff (noninf margin = 0.05 method = score) ;
run; |