[R] Problem with dummy variables controlling for years

Johannes Habel johannes.habel at gmx.de
Mon Mar 2 19:41:56 CET 2009


Hello everybody,

I'm running a panel regression with dummy variables controlling for years.
While this works fine for most of my models, in some cases I get the error
message "In sqrt(diag(object$vcov)) : NaNs wurden erzeugt" and missing
results for some or all of my year dummies:

                                  Estimate Std. Error z-value Pr(>|z|)   
year2000                     -7.48e-02         NA      NA       NA    
year2001                     -6.65e-02         NA      NA       NA    
year2002                     -6.61e-02         NA      NA       NA    
year2003                     -6.49e-02         NA      NA       NA    
year2004                     -5.95e-02         NA      NA       NA    
year2005                     -6.08e-02         NA      NA       NA    

Despite this inconsistency, the other variables as well as my R-quared look
alright.

Can anybody explain me what the reason for this problem is and how it
affects the regression? Are the results still usable in spite of the NaNs?

Thank you very much!

Best,
Johannes


-- 
View this message in context: http://www.nabble.com/Problem-with-dummy-variables-controlling-for-years-tp22294013p22294013.html
Sent from the R help mailing list archive at Nabble.com.




More information about the R-help mailing list