[Limdep Nlogit List] Odp: ordered logit: one threshold missing?
Jadwiga Kostrzewska
dwakotki at wp.pl
Sat Jan 9 22:15:04 EST 2010
Hello Anna-Kaisa,
mu(0)=0 - always, mu(1) and mu(2) are estimated. So this is all right in
your output.
But look at your output, becouse you should use ; Model=Logit not only
;Logit. In my version (NLogit 4)
for:
ORDERED;Lhs=Y;Rhs=X; Logit$
the output is:
(...)
| Model estimated: Jan 09, 2010, 12:10:39 |
| Underlying probabilities based on Normal |
but for:
ORDERED;Lhs=Y;Rhs=X; Model=Logit$
the output is:
(...)
| Model estimated: Jan 09, 2010, 12:12:45 |
| Underlying probabilities based on Logistic |
There is a mistake in a book.
Regards,
Jadwiga Kostrzewska
Dnia 9-01-2010 o godz. 10:10 Anna-Kaisa Kosenius napisał(a):
> Dear All,
>
> I'm estimating the ordered logit model with a 4-level categorical
> variable as a dependent variable (coded 0,1,2,3). However, the output
> only reports two threshold values instead of three.
>
> Has anyone had the same problem before, what could be the reason for
> this, and is there anything that can be done for this?
>
> My command is:
> ORDERED;Lhs=CERTAINT;Rhs=ONE,DIFF,TAX,TAXLOWER,ACCTAX,ACCSCE,TRUSTACT
> ,HIEDUC,GENDER,FREETIME;Logit$
>
> Thanks for help,
> Anna-Kaisa
>
> --
> Anna-Kaisa Kosenius
> Doctoral student
> Department of Economics and Management
> P.O.Box 28 (Koetilantie 5)
> FIN-00014 University of Helsinki
> Tel/Gsm +358 (0)50 4151 219
> anna-kaisa.kosenius at helsinki.fi
> www.mm.helsinki.fi/~kosenius
>
>
>
>
>
> _______________________________________________
> Limdep site list
> Limdep at limdep.itls.usyd.edu.au
> http://limdep.itls.usyd.edu.au
----------------------------------------------------
Kibicuj biało-czerwonym w Zakopanem
Największa impreza sportowa tej zimy!
Zobacz zdjęcia:
http://klik.wp.pl/?adr=http%3A%2F%2Fcorto.www.wp.pl%2Fas%2Fskoki_galeria.html&sid=942
More information about the Limdep
mailing list