C********n 发帖数: 346 | 1 I want to do a joint test for overall difference in a group of variables
between two random samples. These variables include BOTH numerical AND
categorical variables. The null hypothesis is that there is no significant
difference between the two samples in the categorical and numerical variables.
What is the name of the test? Also, if you know how to implement the test
in SAS, it's even better.
Thanks a lot~~ |
w***y 发帖数: 114 | 2 for categorical
%macro cate(var)
proc freq data=
tables &var.*treatment/exact;
output out=pvalue(keep=XP2_fish) exact;
run;
%mend;
for continuous
%cont(var)
proc glm data= outstat=pvalue_con(keep=prob _type_ _name_ where=(type="
ss3"));
class treatment;
model &var.=treatment;
run;
%mend;
you need change it to macro sas code |
C********n 发帖数: 346 | 3 Thank you! So there's no way to do a combined test for both categorical and
numerical variables at the same time?
I suppose that the test for categorical variables in your code is Fisher's
exact test? I know that it can be done for individual variables. Can it be
done jointly for a group of variables?
What about the test for the numerical variables? What is the name of the
test? Is it F-test for means? Is there a test that does not assume normality, or a joint test for difference in median |
b*******r 发帖数: 152 | 4 there will be 'multiple comparison' issue, your type I error be biased. |