[Limdep Nlogit List] Multinomial Probit Model

[Abdulbaki Bilgic]

Ozge,
Oncelikle datalarin dogru set up yapilmamis. Ornegin, Y bagimli
degiskenin hangi secenegini gostermis oldugu anlasilmamaktadir. Bunu
basit bir ornek ile izah edelim: Elimizde 6 kisiye sunulmus olan  tatil
yerleri belirlenmis olsun ve yerler sirasiyla, Mugla, Antalya, Izmir,
Istanbul ve Erzurum olsun. Bu alti kisinin bu yazin yalnizca herhangi
bir ili secmek zorunlulugu olsun ve yapmis olduklari tercihler soyle
olsun:
birinci sahis Antalya, ikinci sahis Erzurum, ucuncu sahis Istanbul,
dorduncu sahis Mugla, besinci sahis Antalya ve son kisi ise Erzurum'u
tercih etmis olsunlar. Birde bu illerin attributesleri vardir ve
sahislarin sosyo-demographic characteristics leri olsun. Sehirlerin
attributesleri nufussayisi, yazin ilin ortalama sicakligi olsun,
kisilerin characteristic degiskenleride yas ve almis olduklari egitim
yillari olsun. Tabi bu attributesleri ve characteristicleri
cogaltabilirsin. Birinci kisi icin data sablonu soyle olmalidir:

Birinci sahis icin:
Sehir         Tercih  Nufussayisi   OrtalamaYazSicaklik    Yas  Egitim
Mugla          0        56,000          40                  40   10
Antalya        1       500,000          45                  40   10
Izmir          0     4,000,000          43                  40   10
Istanbul       0    10,000,000          37                  40   10
Erzurum        0       400,000          30                  40   10

Ikinci sahis icin:
Sehir         Tercih  Nufussayisi   OrtalamaYazSicaklik    Yas  Egitim
Mugla          0        56,000          40                  25   12
Antalya        0       500,000          45                  25   12
Izmir          0     4,000,000          43                  25   12
Istanbul       0    10,000,000          37                  25   12
Erzurum        1       400,000          30                  25   12

Ucuncu sahis icin:
Sehir         Tercih  Nufussay       OYazSic                Yas  Egitim
Mugla          0        56,000          40                  28   18
Antalya        0       500,000          45                  28   18
Izmir          0     4,000,000          43                  28   18
Istanbul       1    10,000,000          37                  28   18
Erzurum        0       400,000          30                  28   18

Dorduncu,besinci ve altinci kisiler icin veri sablonlarini
olusturduktan sonra bu sablonlari birlestirip excelde bir veri setini
olusturuyoruz.
Bundan sonra ise programlama basittir
MNPROBIT;  Lhs=Tercih
        ;  Choices=Mugla,Antalya,Izmir,Istanbul,Erzurum
        ;  Rhs=Nufussay,OYazSic
        ;  rh2=one,yas,egitim
        ;  effects:OYazSic(Antalya)    $
Dikkat etmen gereken bir nokta karakter sayisi 8 gecmemelidir. I hope
this will help you. Take care and if you need assistance, please do not
hesitate to ask for.
Best regards,,,
A.Baki Bilgic
Assistant Professor

--- Özge Akkus <ozgeucar at hacettepe.edu.tr> wrote:

> 
> > Hello Dear Limdep Users,
> 
> > I have an example data set as,
> 
>                    X2             X3              Y
>       0 0 0
>       1 0 1
>       2 0 2
>       3 0 2
>       4 0 1
>       0 1 1
>       1 1 0
>       2 1 2
>       3 1 1
>       4 1 1
>       0 2 0
>       1 2 0
>       2 2 0
>       3 2 1
>       4 2 2
> 
> 
> while Y is the dependent variable with J unordered categories and X2
> and X3
> are independent variables. I would like to apply multinomial probit
> model to
> the data set but I don't know which variables I will attain to
> "Attributes
> (RHS)" and "Characteristics (RH2)" cells. Also, when I applied MPM, 
> an
> error appears such as "1 responses should sum to 1.0. Sum is 0.00". I
> would
> like know the reason of this error.
> 
> I'm waiting for your help.
> 
> > Thanks.
> 
> > Özge
> 
> Hacettepe University
> Faculity of Science
> Department of Statistics
> Beytepe/Ankara-TURKIYE
> 
> _______________________________________________
> Limdep site list
> Limdep at limdep.itls.usyd.edu.au
> http://limdep.itls.usyd.edu.au
> 


Abdulbaki Bilgic, PhD.,
Assistant Professor,
******************************************
******************************************
Current Address:
Harran Üniversitesi, Ziraat Fakültesi,
Tarim Ekonomisi Bölümü, Eyyubiye Kampusu,
63040 Sanliurfa/Turkey
Phone       :+90 414 247 0384/2369
Fax         :+90 414 247 4480
Webpage:http://www.freewebs.com/bilgic
******************************************
******************************************


      ____________________________________________________________________________________
Park yourself in front of a world of choices in alternative vehicles. Visit the Yahoo! Auto Green Center.
http://autos.yahoo.com/green_center/