Page 1 of 1

Estimated coefficient covariance matrix is of reduced rank

Posted: Tue Mar 17, 2015 1:41 am
by u01uw11
Hi! I'm using EViews7.

I've looked through other posts regarding this issue, however, I don't find an explanation for my case. I believe I have more time series observation that cross sections, which could have been one of the issues.

Dependent Variable: LOGTOBINQ Method: Panel Least Squares Date: 03/17/15 Time: 08:27 Sample (adjusted): 2001 2013 Periods included: 13 Cross-sections included: 77 Total panel (unbalanced) observations: 773 White period standard errors & covariance (d.f. corrected) WARNING: estimated coefficient covariance matrix is of reduced rank

I've attached my updated data, and general information about my regression can be found in the link below:

http://forums.eviews.com/viewtopic.php?f=4&t=11742

I get this error message when I include the Age variable. If I leave out the age variable and use only Cross section fixed effects I do NOT get this error message. However, when I use both cross section and period fixed effects, I do get the error message.

I'm not quite sure why this is the case and how to fix it. I would appreciate further insight.

Re: Estimated coefficient covariance matrix is of reduced ra

Posted: Thu Jun 25, 2015 10:05 am
by EViews Glenn
Probably the number of observations, but I can't tell from your description, nor from the workfile.

Please repost including the estimated equations to which you refer.