Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Trouble with mixlogit (error) with binary dependent variable


From   SAM MCCAW <sam2stata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Trouble with mixlogit (error) with binary dependent variable
Date   Mon, 4 Feb 2013 20:50:23 -0500

Thanks much. So if I have one variable, innovation, with a possible
value of (0,1) then my group variable should just be group = 1?

Thanks so much.

SAM

On Wed, Jan 30, 2013 at 6:36 AM, Arne Risa Hole <arnehole@gmail.com> wrote:
> Sam
>
> -mixlogit- is a user-written command available from SJ/SSC.
>
> You've defined the group variable incorrectly - this should be an
> identifier of the choice sets, not the alternatives within a choice
> set. See the examples in
> http://www.stata-journal.com/article.html?article=st0133.
>
> There might be other problems with the way you've set up the data too
> but this is hard to know without seeing a snippet of the data. Have
> you tried running the model using -clogit- first? If this doesn't work
> you can be 100% sure that -mixlogit- won't either.
>
> Arne (author of -mixlogit-)
>
> On 30 January 2013 00:55, SAM MCCAW <sam2stata@gmail.com> wrote:
>> Dear STATA gurus,
>>
>> I am having trouble running a mixed logit model. Please help!
>>
>> - The dependent variable is binary (0,1) (innovation).
>>
>> - My independent variables (var1 var2 var3) are continuous variables
>> (they are logs of a count, excluding 0).
>>
>> -The group variable just takes a value of 1 or 2, because I only have
>> two choice occasions, using: egen group = group(innov_1)
>>
>> - Code: mixlogit innov, group(group) rand(var1 var2 var3) id(id) or
>>
>>          mixlogit innov var4, group(group) rand(var1 var2 var3 var5) id(id)
>>
>> I am getting the following errors:
>>
>> [P]     error . . . . . . . . . . . . . . . . . . . . . . . . Return code 2000
>>         no observations;
>>         You have requested some statistical calculation and there are
>>         no observations on which to perform it.  Perhaps you specified
>>         if exp or in range and inadvertently filtered all the data.
>>
>> (end of search)
>>
>> I have over 30,000 for all the variables specified in the simplest
>> model I ran and I still get the same error.
>>
>> Here are the sumamry statistics for my variables:
>>
>>     Variable |       Obs        Mean    Std. Dev.       Min        Max
>> -------------+--------------------------------------------------------
>> innov_1       |     32523    .4774467    .4994988          0          1
>> var1            |     33065    2.104231    1.589854          0   4.859812
>> var2            |     33065    1.669999      1.0031            0   4.060443
>> var3            |     33065    .6736071    .9022467          0    3.78419
>> var4            |     33065    2.141496    2.795201          0   9.953027
>> -------------+--------------------------------------------------------
>> var5            |     28893    1.506202    .9606917          0   3.295837
>> var6            |      5953   -2.335463    1.106637   -4.60517    1.96611
>>
>> I am completely lost about what to do next.
>>
>> Thanks for your help.
>>
>> SAM
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/faqs/resources/statalist-faq/
>> *   http://www.ats.ucla.edu/stat/stata/
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index