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

From |
"Schaffer, Mark E" <[email protected]> |

To |
<[email protected]> |

Subject |
RE: Repost: instrumental variables regression with random effects GLS using cross-section data and endogenous binary independent variable |

Date |
Mon, 8 Oct 2007 16:42:30 +0100 |

```
Renuka,
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Renuka Metcalfe
> Sent: Monday, October 08, 2007 9:50 AM
> To: [email protected]; [email protected]
> Cc: [email protected]
> Subject: Re: Repost: instrumental variables regression with
> random effects GLS using cross-section data and endogenous
> binary independent variable
>
> What I mean is I estimated initially a random effects GLS
> model; now I want to correct for the endogenity between the
> dependent and one of the independent variable which is a
> binary variable as stated in my email below.
Why not use -xtivreg-? It does IV random effects in two flavours (G2SLS
and EC2SLS), both of which are consistent whether or not the endogenous
regressor is binary or continuous.
HTH,
Mark
> --- [email protected] wrote:
>
> > I don't know what "ran random effects GLS" means, since you declare
> > having a cross-section. -ivreg- (and related commands, such as
> > -ivreg2-, -xtivreg-,
> > etc.) can be used with a dummy endogenous variable, too.
> >
> > Nicola
> >
> > At 02.33 05/10/2007 -0400, Renuka Metcalfe wrote:
> > >Dear Statalisters
> > >
> > >I have a cross-section data and my dependent
> > variable
> > >is continuous and I have one independent variable which is
> endogenous
> > >and is binary. I have an instrument for the endogenous binary
> > >independent variable. I ran random effects GLS as follows
> > >
> > >Y x1 x2 x3 x4
> > >
> > >Since x4 is endogenous with Y, I now want to use an
> instrument Z. Y
> > >is a continuous variable x4 is an endogenous binary variable. The
> > >data are also
> > grouped
> > >across workplaces.
> > >
> > >Looking at the archives it was suggested that a
> > probit
> > >be used for the first stage estimation of the independent
> endogenous
> > >binary variable. But the dependent variable there was censored.
> > >
> > >If you need further information please let me know.
> > >
> > >Thanks in advance
> > >
> > >Best wishes
> >
> >
>
>
>
> ____________________________________________________
> One email account is enough. Simplify your life by switching
> other accounts into Yahoo! Mail.
> http://uk.docs.yahoo.com/mail/trueswitch_overview
> *
> * 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/
>
*
* 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/
```

**References**:

- Prev by Date:
**Re: st: Connecting baseline dataset to visit dataset** - Next by Date:
**Re: st: Which weight** - Previous by thread:
**Re: Repost: instrumental variables regression with random effects GLS using cross-section data and endogenous binary independent variable** - Next by thread:
**st: DHS ten years of data** - Index(es):

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