Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: Tobit in Stata vs. Limdep


From   "Scott Merryman" <smerryman@kc.rr.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Re: Tobit in Stata vs. Limdep
Date   Mon, 7 Jul 2003 22:03:22 -0500

----- Original Message ----- 
From: <frone@ria.buffalo.edu>
To: <statalist@hsphsun2.harvard.edu>
Sent: Monday, July 07, 2003 8:47 PM
Subject: st: Tobit in Stata vs. Limdep


> I forgot to mention that I'm using windows versions of both Stata/SE 8 and
> Limdep 8.
> ----------------------
> I've run a tobit analysis using Stata and Limdep.  However, the
> coefficients (and z-tests) are quite different.  Has any one done such a
> comparison and can shed light on why this might happen?  Just as a
> secondary check, I dichotomized the outcome and ran a probit in both
> programs and obtained Identical results.
>
> Mike Frone
>

Mike,

I do not have access to Limdep 8, however I can produce the same Tobit
results with Stata 8 and Limdep 7.

Scott-

Example:

. use http://www.ats.ucla.edu/stat/stata/webbooks/reg/acadindx
(max possible on acadindx is 200)

. tobit acadindx female reading writing, ul(200)

Tobit estimates                                   Number of obs   =
200
                                                  LR chi2(3)      =
190.39
                                                  Prob > chi2     =
0.0000
Log likelihood = -718.06362                       Pseudo R2       =
0.1171

----------------------------------------------------------------------------
--
    acadindx |      Coef.   Std. Err.      t    P>|t|     [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
      female |  -6.347316   1.692441    -3.75
0.000    -9.684943   -3.009688
     reading |   .7776857   .0996928     7.80   0.000     .5810837
.9742877
     writing |   .8111221    .110211     7.36   0.000     .5937773
1.028467
       _cons |   92.73782   4.803441    19.31   0.000     83.26506
102.2106
-------------+--------------------------------------------------------------
--
         _se |   10.98973   .5817477           (Ancillary parameter)
----------------------------------------------------------------------------
--

  Obs. summary:        184     uncensored observations
                        16 right-censored observations at acadindx>=200




 TOBIT;Lhs=ACADINDX;Rhs=ONE,FEMALE,READING,WRITING;Limit=200;Upper$


            +---------------------------------------------+
              | Limited Dependent Variable Model - CENSORED |
              | Maximum Likelihood Estimates                |
              | Dependent variable             ACADINDX     |
              | Weighting variable                  ONE     |
              | Number of observations              200     |
              | Iterations completed                  4     |
              | Log likelihood function       -718.0636     |
              | Threshold values for the model:             |
              | Lower=-infinity      Upper=  200.0000       |
              +---------------------------------------------+
+---------+--------------+----------------+--------+---------+----------+
|Variable | Coefficient  | Standard Error |b/St.Er.|P[|Z|>z] | Mean of X|
+---------+--------------+----------------+--------+---------+----------+
          Primary Index Equation for Model
 Constant  92.73781960      4.8034407       19.307   .0000
 FEMALE   -6.347315691      1.6924410       -3.750   .0002  .54500000
 READING   .7776857346      .99692772E-01    7.801   .0000  52.230000
 WRITING   .8111221138      .11021102        7.360   .0000  52.775000
         Disturbance standard deviation
 Sigma     10.98973474      .58174773       18.891   .0000


*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index