Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: clogit interpretation


From   Julia Gamas <[email protected]>
To   [email protected]
Subject   st: clogit interpretation
Date   Fri, 5 May 2006 08:39:44 -0700 (PDT)

Hi all,
I'm trying to make sure that my interpretation of my
clogit coefficient results is correct.  My basic
question is: do the signs reported in the clogit refer
to the impact that a given variable has on overal
utility OR on the utility of one choice as compared to
the other?
I have two choices: "used private transportation" or
"used public transit".  The individual specific
variables have been "dummied" such as:
genderprivate,ageprivate incomeprivate.  
I also have a choice specific variable: "trip
duration".
1.  My first question is if I get a positive
coefficient on my choice dependent variable (trip
time), does this mean that the longer the trip, the
more likely I am to choose private transportation? 
But in essence, shouldn't this coefficient be negative
(as trip time takes away from utility in general)?  
2.  Furthermore, if (following Ben-Akiva 1996) I
assume that trip times have a different quality for
public than for private transit, then should the
coefficient for publictime be positive (the more trip
time on public transit, the more likely I am to
drive)?  And should the coefficient on private trip
time be negative?  Or should both be negative?
Examples of 1 follows:

1.  Variable "duracion" is the trip duration.
. clogit choice private incpppriv pphhpriv headpriv
publiccost privatecost duracion, group (numob)
note: 393 groups (393 obs) dropped due to all positive
or
      all negative outcomes.

Iteration 0:   log likelihood = -3513.9953  
Iteration 1:   log likelihood = -26.537414  
Iteration 2:   log likelihood =          0  
Iteration 3:   log likelihood =          0  

Conditional (fixed-effects) logistic regression  
Number of obs   =      83102
                                                  LR
chi2(7)      =   57601.92
                                                  Prob
> chi2     =     0.0000
Log likelihood =          0                      
Pseudo R2       =     1.0000

------------------------------------------------------------------------------
      choice |      Coef.   Std. Err.      z    P>|z| 
   [95% Conf. Interval]
-------------+----------------------------------------------------------------
     private |  -41.57196   1.80e+07    -0.00   1.000 
  -3.53e+07    3.53e+07
   incpppriv |  -.0033245   68243.68    -0.00   1.000 
  -133755.2    133755.2
    pphhpriv |   .0100637    2625669     0.00   1.000 
   -5146218     5146218
    headpriv |  -.3225972   1.02e+07    -0.00   1.000 
  -2.00e+07    2.00e+07
  publiccost |  -82.60774   1.31e+07    -0.00   1.000 
  -2.57e+07    2.57e+07
 privatecost |  -.0002418   270797.4    -0.00   1.000 
  -530753.1    530753.1
    duracion |  -.0072196   192486.6    -0.00   1.000 
  -377266.9    377266.9
------------------------------------------------------------------------------

Any help will be appreciated.

Sincerely,

Julia A. Gamas, Ph.D.
Mario Molina Center for Energy and Strategic Studies


END OF CLOGIT INTERPRETATION QUESTION **************

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index