Carlos,
You could also possibly use a fixed effects vector decomposition procedure,
see Plumper, V., Troeger, V. E., 2007. Efficient estimation of
time-invariant and rarely changing variables in finite sample panel analyses
with unit fixed effects. Political Analysis, 15, 124-139.
Alternatively, if want to estimate random effects, the Hausman-Taylor
estimator might be useful to solve the problem Dan mentioned (i.e.
regressors correlated with unit-specific error terms).
Best,
Johan
> -----Original Message-----
> From: Political Methodology Society [mailto:[log in to unmask]] On
> Behalf Of Carlos Rodriguez
> Sent: Monday, January 19, 2009 5:46 PM
> To: [log in to unmask]
> Subject: [POLMETH] Fixed-effects and slow-moving variables
>
> Dear list participants,
>
> I have a basic question I'd be grateful if sb could clarify: when
> using a coutnry FE model with TSCS data, should we one LEAVE OUT all
> control variables that do not change much over time? Should one
> INCLUDE them but refrain from interpreting their coefficients? I
> guess my question would be whehter including such slow-moving
> variables would bias or otherwise affect the estimates for the
> variables that do change over time. Also, how much should a variable
> change over time to be fine to interpret it with FE and how to
> ascertain the magnituted of this change?
>
> Thanks, as usual, and HAPPY 2009
>
> **********************************************************
> 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
>
> **********************************************************
**********************************************************
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
**********************************************************
|