Re: st: Marginal Effects: CLogit

[Wameq Raza <wameq.r@gmail.com>]

Dear Nick,
Thank you very much for the clarification, I'm not sure how I missed
that. As I previously mentioned,
I'm just trying to get the marginal effects of the variables from the
estimated model.
Am I going about this the wrong way?
Wameq

On Sat, Oct 6, 2012 at 7:09 PM, Nick Cox <njcoxstata@gmail.com> wrote:
> The help for -margins- starts off
>
> margins [marginlist] [if] [in] [weight] [, response_options options]
>
>     where marginlist is a list of factor variables or interactions that
>     appear in the current estimation results.
>
> That means what it says: so, you are asking something -margins- can't
> do it if you offer up continuous variables _as if_ they were factor
> variables. That just won't work.
>
> The real questions are thus exactly what you want out of -margins-;
> whether that makes sense, and how else to get it, on which others can
> advise better than I can.
>
> Nick
>
> On Sat, Oct 6, 2012 at 5:51 PM, Wameq Raza <wameq.r@gmail.com> wrote:
>
>> I was wondering if someone could kindly help me figure out how to get
>> the marginal effects of my conditional logit model.
>> The syntax I'm using:
>>
>> clogit hw cost_ time_ wait_,group(id)
>> margins cost_ time_ wait_, predict(xb)
>>
>> These are logged alt specific variables, but the error STATA keeps on
>> throwing back at me is:
>> "cost_:  factor variables may not contain noninteger values"
>>
>> However, these are continuous variables and not factor vars. I'm not
>> sure how exactly to proceed.
> *
> *   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/



-- 
W A M E Q   R A Z A
*
*   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/