Page 1 of 1
panel data regression with exponential form
Posted: Fri Jun 22, 2012 3:21 am
by Sima
Hi
I want to know how can I run a panel data regression with exponential form? I've been told that the formula ls y = c(1) + exp(c(2)*x) doesn't work for panel data.Is it correct?
panel data regression with exponential form
Posted: Fri Jun 22, 2012 6:32 am
by EViews Gareth
What is the form of the equation you're trying to estimate?
Re: panel data regression with exponential form
Posted: Fri Jun 22, 2012 10:52 am
by EViews Glenn
And what is your error specification and how does the panel nature of the data enter into the spec.
Re: panel data regression with exponential form
Posted: Fri Jun 22, 2012 9:57 pm
by Sima
What is the form of the equation you're trying to estimate?
My equation is u=e^av where u is unemployment rate and v is vacancy rate.I ran u=c(1)+exp(c(2)*v) and it worked without any error .but my teacher said that it's not correct because it doesn't consider the fixed and random effect.I don't understand why?
panel data regression with exponential form
Posted: Fri Jun 22, 2012 10:14 pm
by EViews Gareth
Did you include fixed or random effects?
Re: panel data regression with exponential form
Posted: Fri Jun 22, 2012 10:33 pm
by Sima
Did you include fixed or random effects?
yes,I included fixed effects.
Re: panel data regression with exponential form
Posted: Mon Jun 25, 2012 10:30 am
by EViews Glenn
Any fixed effects added to the specification will be as a linear term added to the end of your nonlinear specification. Estimation is done in the usual fashion for fixed effects by iterating the linearized specification to convergence.