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

Re: st: Command for Two-limit tobit model

From   Maarten buis <>
Subject   Re: st: Command for Two-limit tobit model
Date   Thu, 10 Apr 2008 08:32:10 +0100 (BST)

--- Marco Velįstegui Andrade <> wrote:
> Is someone knows the commands to estimate a two-limit tobit model,
> its marginal effects and to test for heteroscedasticity.

The command is called -intreg-, and you can get marginal effects using
-mfx-, notice that there are multiple types of marginal effects,
depending whether you want the effect on the latent variable, the
observed variable. The first -mfx- command gives the marginal effects
on the latent variable (and is superfluous as will always give the same
estimates as -intreg-), and the second -mfx- command gives you the
marginal effect on the observed variable. I don't know any test for
heteroscedasticity in this context, but I do know that you should first
make certain that the effects of your explanatory variable have the
correct functional form.

-- Hope this helps,

*---------------- begin example --------------------------
sysuse nlsw88, clear
gen ln_w = ln(wage)

// create a new variable
// censored from below at 1.15
// censored from above at 2.5

gen lb = cond(ln_w < 1.15, . , /// lower lb is -infty
         cond(ln_w > 2.5, 2.5, /// upper lb is 2.5
         ln_w))                /// in between ln_w

gen ub = cond(ln_w < 1.15, 1.15 , /// lower ub is 1.15
         cond(ln_w > 2.5, .,      /// upper ub is infty
         ln_w))                   /// in between ln_w

intreg lb ub ttl_exp tenure grade age union
mfx, predict(ystar(1.15,2.5))
*------------------ end example --------------------------
(For more on how to use examples I sent to the Statalist, see )

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Yahoo! For Good helps you make a difference
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index