enquiry about using likelihood ratio test when employing conditional logistics regression !

[Danah Abdul]

Hello,
I am testing a categorical variable (ethnicity) with a binary outcome (death). I want to obtain a single P value across all categories so I employed the likelihood ratio test for that purpose. but when comparing the null hypothesis model, I don't get any odds ratio (empty space). Yet, I end up with a P value.

I use the following commands:

clogit death i.ethnicity, group(matchedid) or
store est Danah
clogit death if ethnicity !=. , group(matchedid) or
lrtest Danah

it is after this command that I get no odds ratio:

clogit death if ethnicity !=. , group(matchedid) or

Nevertheless, I end up with a single P value after

lrtest Danah

When the same commands are run for "logistic" (unconditional logistic regression) I do get odds ratio testing the null hypothesis model, which made me suspect that I am doing something incorrect.

Can anyone thankfully help me on the correct way to run a likelihood ratio test for variables tested with conditional logistic regression.

[Danah Abdul]

Hello,
For those reading the post,
I realised that the Y intercept (_cons) is non-existent for conditional logistic regression, and hence I get empty observations whilst conducting the likelihood ratio test as explained above.

Thanks
Danah