How to use mixlogit command for DCE?

[Nabittun Nahar]

Hi there,

#Stata18

I have some pretest samples for a discrete choice experiment. I want to fit the mixed logit model following Hole's (2007) to capture the heterogeneity of preferences; however, I cannot make the mixlogit command work.

The error I got from mixlogit y price, rand(pesticide type) group(id) id(group) nrep(50) is "command mixlogit is unrecognized" r(199). Also, when I try to get help from help mixlogit, it directs me to search mixlogit. My colleague uses a previous stata version, and mixlogit works fine for him.

Here is an example of the data.
----------------------- copy starting from the next line -----------------------

* Example generated by -dataex-. For more info, type help dataex
clear
input byte(choice1 pesticide) int price byte(type group)
0 1 102 2 7
0 1  93 2 3
0 1 148 3 4
0 2  84 1 3
0 2 124 3 7
0 2 129 2 4
1 3  77 1 4
1 3 160 3 3
1 3  98 1 7
0 1  93 2 3
0 1 102 2 7
0 1 148 3 4
0 2  84 1 3
0 2 124 3 7
0 2 129 2 4
1 3 160 3 3
1 3  98 1 7
1 3  77 1 4
0 1 148 3 4
2 1 102 2 7
3 1  93 2 3
0 2  84 1 3
0 2 129 2 4
0 2 124 3 7
1 3  77 1 4
0 3 160 3 3
0 3  98 1 7
0 1 102 2 7
0 1 148 3 4
0 2 124 3 7
0 2 129 2 4
1 3  98 1 7
1 3  77 1 4
0 1 102 2 7
0 1 148 3 4
0 1  93 2 3
0 2 129 2 4
2 2  84 1 3
0 2 124 3 7
0 3 160 3 3
1 3  98 1 7
1 3  77 1 4
0 1 102 2 7
0 1 148 3 4
0 1  93 2 3
0 2 129 2 4
0 2 124 3 7
2 2  84 1 3
1 3  77 1 4
0 3 160 3 3
1 3  98 1 7
0 1  77 1 9
0 1 129 2 6
0 1  84 1 1
1 2 102 2 9
0 2 148 3 6
3 2  93 2 1
0 3 160 3 9
3 3  77 1 6
0 3 140 3 1
0 1 124 3 2
0 1 160 3 8
0 1  98 1 5
2 2  98 1 2
2 2  77 1 8
0 2 129 2 5
1 3 148 3 5
0 3 102 2 2
0 3 102 2 8
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
. .   . . .
end

where choice is the dependent variable. Price, pesticide, and type are attributes. Group is for choice sets.
I am not done with the data entry yet. Before I enter all the data into Excel, I wanted to check if Mixlogit works for me. I suspect my data is probably not in the right format for mixlogit.
Moreover, I tried to follow cmmixlogit, but I'm unsure about the casevar. Is it the respondent's ID or the choice set?
Any help is much appreciated. Thank you.

[Nabittun Nahar]

Update about the mixologist command!
Please ignore the part about mixlogit. I have a solution to the mixlogit command.

[Nabittun Nahar]

Sorry about the typo.