[Limdep Nlogit List] question about defining universal choice set when all choices aren't available to everyone

[William Greene]

Dear Kristy. I've received your query about nlogit.  It appears there is an
issue
with the coding of NIJ and ALTIJ.  I'm working on the program. I'll let
you know
what I find.
/Bill Greene

On Fri, Apr 7, 2023 at 7:02 PM Kristy Wallmo - NOAA Federal via Limdep <
limdep at mailman.sydney.edu.au> wrote:

> Dear Dr. Greene,
>
> I am trying to estimate a location choice model for fishermen and am
> following the example you provide in the Nlogit documentation where not
> everyone has every alternative available to them.  In my case everyone’s
> choice set contains 15 alternatives, and the universal set of alternatives
> is 71 sites.  I am starting with a very simple model where site choice is
> function of expected revenue( bigeye_o), trip cost (tripcost) and other
> boats (boats5x5)
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> My code is the following:
>
>
>
> NLOGIT ; Lhs = chosen,  nij,  altij
>
>     ; Choices =
>
> 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71
>
>     ; Rhs = bigeye_o, boats5x5, tripcost
>
>     ; check data $
>
>
>
> nij is 15 for everyone, and altij is a suite of 15 different site numbers.
> I have defined the universal choice set as both numbers 1- 71 and tried
> using generic labels e.g. a, b, c,..etc.  When I use the check data I get
> the errors below, and my first question is about the missing values:  I
> have checked and double checked and I do not have any missing data in the
> dataset, and all data is numeric.  Second, within each set of the 15 choice
> alternatives there are no duplicates, so I am not quite understanding the
> bulk of the error messages.  Any advice you have would be much appreciated.
>  – thanks Kristy
>
> +----------------------------------------------------------+
>
> | Inspecting the data set before estimation.               |
>
> | These errors mark observations which will be skipped.    |
>
> | Row Individual = 1st row then group number of data block |
>
> +----------------------------------------------------------+
>
>     0   2270  Missing value found for characteristic or attribute in
> utility
>
>     0   2540  Universal choice set, ID var. (3rd LHS) takes same value
> twice
>
>     0   2541  Universal choice set, ID var. (3rd LHS) takes same value
> twice
>
>     0   2542  Universal choice set, ID var. (3rd LHS) takes same value
> twice
>
>     0   2543  Universal choice set, ID var. (3rd LHS) takes same value
> twice
>
>     0   2544  Universal choice set, ID var. (3rd LHS) takes same value
> twice
>
>     0   2545  Universal choice set, ID var. (3rd LHS) takes same value
> twice
>
> 38179   2546  Universal choice set, ID var. (3rd LHS) takes same value
> twice
>
>     0   2547  Universal choice set, ID var. (3rd LHS) takes same value
> twice
>
>   I shortened this but basically 196 error lines that say this…
>
>     0   8690  Missing value found for characteristic or attribute in
> utility
>
>
>
> Normal exit:   5 iterations. Status=0, F=    17957.48
>
>
>
>
> -----------------------------------------------------------------------------
>
> Discrete choice (multinomial logit) model
>
> Dependent variable               Choice
>
> Log likelihood function    -17957.48310
>
> Estimation based on N =   8768, K =   3
>
> Inf.Cr.AIC  =  35921.0 AIC/N =    4.097
>
> Model estimated: Apr 07, 2023, 12:25:22
>
> R2=1-LogL/LogL* Log-L fncn R-sqrd R2Adj
>
> Constants only must be computed directly
>
>                Use NLOGIT ;...;RHS=ONE$
>
> Response data are given as ind. choices
>
> Number of obs.=  8964, skipped  196 obs
>
>
> --------+--------------------------------------------------------------------
>
>         |                  Standard            Prob.      95% Confidence
>
>   CHOSEN|  Coefficient       Error       z    |z|>Z*         Interval
>
>
> --------+--------------------------------------------------------------------
>
> BIGEYE_O| .79760D-05***   .6999D-06    11.40  .0000  .66042D-05  .93478D-05
>
> BOATS5X5|     .06895***      .00108    63.89  .0000      .06683    .07106
>
> TRIPCOST|    -.00031***   .8394D-05   -37.20  .0000     -.00033   -.00030
>
>
> --------+--------------------------------------------------------------------
>
> Note: nnnnn.D-xx or D+xx => multiply by 10 to -xx or +xx.
>
> Note: ***, **, * ==>  Significance at 1%, 5%, 10% level.
>
>
> -----------------------------------------------------------------------------
>
> --
> Kristy Wallmo, Ph.D.
> Economist, National Marine Fisheries Service
> Silver Spring, MD
> _______________________________________________
> Limdep site list
> Limdep at mailman.sydney.edu.au
> http://limdep.itls.usyd.edu.au
>


-- 
William Greene
Department of Economics, emeritus
Stern School of Business, New York University
URL: https://protect-au.mimecast.com/s/Yf5sCOMKzVTNN7ZlRtEYgIm?domain=people.stern.nyu.edu
Email: wgreene at stern.nyu.edu
Editor in Chief: Foundations and Trends in Econometrics
Associate Editor: Economics Letters
Associate Editor: Journal of Business and Economic Statistics