Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: bootstrap two-stage logit

From   Austin Nichols <>
Subject   Re: st: bootstrap two-stage logit
Date   Wed, 11 May 2011 12:08:00 -0400

Why are you worried about estimating SEs for an inconsistent estimator?
Do you have a ref that advocates your "two-stage IV logit estimation"
(see comments on Grogger by googling "grogger gmm logit iv")?

On Wed, May 11, 2011 at 10:41 AM, xueliansharon <> wrote:
> Hi,
> I want to estimate a logit model with one continuous endogenous explanatory
> variable (educ), I have two instruments (moth and fath) in hand to address
> the endogeneity, so I would like to use two-stage IV logit estimation. I
> want to use bootstrap to return the point estimates of the two-stage logit
> model and their bootstrap standard errors.
> The following are my codes, but the returned bootstrapped standard errors
> are the same to the SEs of second-stage logit model (which is wrong and
> needs to corrected by bootstrap), i.e. the bootstrap didn't make any
> difference in correcting the SEs of second-stage logit model. Can anyone
> help me to figure out the mistake in the program? Many thanks!!!
> use, clear
> capture program drop myprog
> prog myprog, eclass
>  regress educ age kidslt6 kidsge6 city moth fath
>  capture drop peduc
>  predict peduc,xb
>  logit inlf peduc age kidslt6 kidsge6 city
>        tempvar touse
>        tempname b V
>        matrix `b'=e(b)
>        matrix `V'=e(V)
>        quietly gen byte `touse' = e(sample)
>        ereturn post `b' `V', esample(`touse')
>        ereturn local cmd "myprog"
> end
> bootstrap, reps(20) seed(1014): myprog
> Regards,
> Sharon

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index