Subject: | |
From: | |
Reply To: | |
Date: | Sat, 6 Dec 2008 19:52:20 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Carlos,
I would use the instrument for which the IV assumptions are the most
plausible. Moreover, overidentification tests are of limited usefulness
because failing to reject the null is not the same as confirming the
null. Moreover, they also assume that treatment effects are the same for
all units which is often not terribly plausible.
cheers,
Holger
Carlos Rodriguez wrote:
> Hi everyone,
>
> Say I have two IVs but they are very highly correlated between
> themselves. Is it still useful to use both to instrument for the same
> endogenous regressor and then run overidentification tests, or should
> I just choose one of them given to intrument for the endogenous
> regressor (given that the two IVs are so highly correlated...perhaps
> this causes problesms?)?
>
> Thank you
>
> Good evening,
> Carlos Rodriguez
>
> **********************************************************
> Political Methodology E-Mail List
> Editors: Melanie Goodrich, <[log in to unmask]>
> Delia Bailey, <[log in to unmask]>
> **********************************************************
> Send messages to [log in to unmask]
> To join the list, cancel your subscription, or modify
> your subscription settings visit:
>
> http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************
>
--
Holger Lutz Kern, PhD
Postdoctoral Associate
Yale University
The MacMillan Center
Program on Democracy
34 Hillhouse Avenue
New Haven, CT 06520
http://pantheon.yale.edu/~hlk8/
**********************************************************
Political Methodology E-Mail List
Editors: Melanie Goodrich, <[log in to unmask]>
Delia Bailey, <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
|
|
|