EViews.com

hi, i have simple question
how to add a linear restriction in a model

example: cobb douglas production function f(K,L)=A*[K^c(2)]*[L^c(3)]
y=f(K,L); A=c(1); log(K)=x; log(L)=z
y=c(1)+c(2)*x+c(3)*z

1/i would like to restricted c(3)=0,2 at first and show c(1) c(2) c(3) in output equation
2/second restriction c(2)+c(3)=1 and show c(1) c(2) c(3) in output equation

have you any idea

hi, i have simple question
how to add a linear restriction in a model

example: cobb douglas production function f(K,L)=A*[K^c(2)]*[L^c(3)]
y=f(K,L); A=c(1); log(K)=x; log(L)=z
y=c(1)+c(2)*x+c(3)*z

1/i would like to restricted c(3)=0,2 at first and show c(1) c(2) c(3) in output equation
2/second restriction c(2)+c(3)=1 and show c(1) c(2) c(3) in output equation

have you any idea

There isn't a way to "show" the restricted coefficients in the output, but you can do the estimation by
y = c(1)+c(2)*x+0.2*z

and
y = c(1)+c(2)*x+(1-c(2))*z

If you then do a Wald test with the expression for the restricted value, EViews will give you the equivalent t-statistics...

thank's for your help and your reply MR startz and MR Glenn.