OLS Regression with restraints on coeffcients

[thijs]

Hi,

I have a problem with Eviews to which I can't seem to find the answer in the help files.

I have the following OLS equation set up:

eq1.ls Y = c(1) + c(2)*A + c(3)*B + c(4)*D + c(5)*E + c(6)*F

Now I want to run this regression on my data, however with a constraint that c(2)+c(4)+c(5)+c(6) = 1.

How can I process this constraint?

Any help is much appreciated, thanks.

Thijs

[startz]

Hi,

I have a problem with Eviews to which I can't seem to find the answer in the help files.

I have the following OLS equation set up:

eq1.ls Y = c(1) + c(2)*A + c(3)*B + c(4)*D + c(5)*E + c(6)*F

Now I want to run this regression on my data, however with a constraint that c(2)+c(4)+c(5)+c(6) = 1.

How can I process this constraint?

Any help is much appreciated, thanks.

Thijs

EViews does not perform constrained regression. However, for linear constraints you can always impose the constraint by substitution. For example

Code: Select all

eq1.ls Y = c(1) + c(2)*A + c(3)*B + c(4)*D + c(5)*E + (1-c(2)-c(3)-c(4)-c(5))*F

[thijs]

ok thanks