conditional logit model

[samaneh khaef]

Hello
I am trying to use conditional logit model in order to see why people go to a specific neighborhood. to do so I have people characteristics (age and gender which is as1 and 2) and also the place characteristics(the employment rate).
I used the following syntax:

asclogit choice employment, casevars(gender age) case(id) alternatives(place)
a sample of my data is :

id	choice	gender	age	employment	place
1	0	0	0	0.16	1
1	0	0	0	0.3	2
1	1	2	31	0.15	3
2	1	1	40	0.16	1
2	0	0	0	0.3	2
2	0	0	0	0.15	3
3	0	0	0	0.16	1
3	1	1	23	0.3	2
3	0	0	0	0.15	3
4	1	2	45	0.16	1
4	0	0	0	0.3	2
4	0	0	0	0.15	3
5	0	0	0	0.16	1
5	0	0	0	0.3	2
5	1	1	30	0.15	3

but I got this message:
"no cases remain after removing invalid observation"

can anyone help me?

[Maarten Buis]

There is something wrong with your data. So we need to know exactly what your data looks like. If you type down a sample of your data, then you give us what you think your data looks like. The fact that you have a problem, tells us that the two are not the same. The way give us an exact sample from your data is for you to use dataex, and paste the resulting output in the post.

[samaneh khaef]

Dear Maarten
Thanks for your comment. since I am working with register data I am not allowed to extract them. that is why I have not added them here. I try to find a way for extraction and come back
many thanks

[samaneh khaef]

I checked my data. I had missing instead of zero in gender. now I fix it. but bow I get this error that"gender is not constant with each case". It is so weird since I check for cases and it is either zero or Femal\Male.