R help - Jul 2011 - Fixed effects using Within transformation in PLM package

Hi all,

I am trying to do my own fixed effects regression using the Within function
in PLM. I apply the Within function to all my pseries and then run OLS on
the transformed vectors using lm().

When I compare the results to those obtained via plm ("within"), the
estimates are not always the same. Specifically, if there are missing values
(NA), the parameter estimates are not the same. I am aware that standard
errors and degrees of freedom have to be corrected. However, as far as I
know the parameter estimates should be the same.

Is R doing any specific correction for the missing values?

I really appreciate any comments on this.

Regards,

Luca

	[[alternative HTML version deleted]]

Dear R community,

I am trying to do my own fixed effects regression using the Within function
in PLM. I apply the Within function to all my pseries and then run OLS on
the transformed vectors using lm().

When I compare the results to those obtained via plm ("within"), the
estimates are not always the same. Specifically, if there are missing values
(NA), the parameter estimates are not the same. I am aware that standard
errors and degrees of freedom have to be corrected. However, as far as I
know the parameter estimates should be the same.

Is R doing any specific correction for the missing values?

I really appreciate any comments on this.

Regards,

Luca

	[[alternative HTML version deleted]]