Skip to content
Prev 314677 / 398502 Next

R: plm random effect: the estimated variance of the individual effect is negative

Dear Matteo,
as I stated in my previous email, if the random effects variance is set to zero then they are identically zero (as mean=0, sd=0) hence you just need to use 'plm(<...> , model="pooling")'.
Best,
Giovanni

-----Messaggio originale-----
Da: matteo ognibene [mailto:ognibenematteo at hotmail.it] 
Inviato: venerd? 4 gennaio 2013 20.08
A: Millo Giovanni; r-help at r-project.org
Oggetto: RE: [R] plm random effect: the estimated variance of the individual effect is negative


Thanks Giovanni,

but I already tried with the others "random.method" and it doesn't work anyway.
Do you think is it possible change the if statement in the plm formula, to set the value to 0 if sigma<0 ?
Because if I insert a dummy variabile in the formula R gives me this error:
Thanks
Matteo Ognibene


----------------------------------------