R help - May 2011 - System is computationally singular error for plm random effects models

Dear all,

I am using the plm package for both fixed and random effects models on my
country-year panel data. However, for some of the random effects models I
get the following error:

Error in solve.default(OM) :
  system is computationally singular: reciprocal condition number 1.78233e-18

The same models work fine for fixed effects. I have also noticed that once I
remove some of my variables (mainly the Year variable) from the models, the
regression can be completed.

Does anyone know what the reason for this error might be? Any commenst will
be highly appreciated and please accept my apologies if my question is too
basic.

Kind regards,
Stanislav

	[[alternative HTML version deleted]]