| |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Re: doubt on panel data estimation
Rodrigo A. Alfaro wrote:
Dear M Parameswaran
There is a "new" literature about estimating fixed-effects in nonlinear
models. Several paper of J. Hahn, but I recommend to start with
a good survey by M. Arellano and J. Hahn (2006) "Understanding
Bias in Nonlinear panel Models" (should be available on-line).
Rodrigo.
The paper that Rodrigo refers to is available here:
http://cemmap.ifs.org.uk/papers/vol3_chap12.pdf
Dave Harless
----- Original Message -----
From: "M.Parameswaran" <[email protected]>
To: <[email protected]>
Sent: Tuesday, June 27, 2006 7:36 AM
Subject: st: doubt on panel data estimation
Dear Friends,
I am a researcher at Centre for Development Studies, Kerala, India. I
have a panel data on manufacturing firms for six years. I want to
estimate a tobit model using this data. I understand that consistent
estimation of random effect tobit model requires strict exogeneity of
the regressors, which is not valid in my data set; current value of
the depended variable affects future value of regressors. Another
option is to use pooled estimation that ignore firm's specific
effects. I would be highly grateful if you could give your advice on
the choice between these two estimates or some other better methods.
Looking forward to your response,
Thanking you,
M Parameswaran
*
* 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/