Re: st: Fixed effect model with 4,000 panels for 400 periods of timeand clustering by panel?

[smerryman@kc.rr.com]

Creating 400 dummy variables in a dataset of 1,600,000 observations is 
going to take a lot memory.  The FAQ "Approximizing the size of a 
dataset"  at http://www.stata.com/support/faqs/data/howbig.html might 
be helpful.

Rather than treating each time period as a separate effect, would it 
be possible to create year and month dummy variables.  This would only 
require adding 34 year and 12 month dummies.

Scott


----- Original Message -----
From: Julian Cristia <Julian.Cristia@cbo.gov>
Date: Wednesday, February 8, 2006 10:47 am
Subject: st: Fixed effect model with 4,000 panels for 400 periods of 
time	and clustering by panel?

> Hi all.
> 
> I have Intercooled Stata 8.0. I am trying to estimate a fixed-
> effects model with 4,000 panels through 400 periods of time. I 
> want to add dummies by panels and time. I also want to cluster 
> observations by panels.
> 
> I tried with:
> 
> xi: areg work i.month anychildren pregnant [pweight=weight], 
> absorb(panelid) cluster(panelid)
> 
> but Stata gives me the following error message
> 
> "no room to add more variables due to width
> An attempt was made to add a variable*"
> 
> Any suggestions? Thanks!!!
> 
> Julian Cristia
>
*
*   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/