Comparing coefficients accross regressions
Posted: Fri Jan 19, 2018 10:18 am
Hello Eviews users and moderators
Please assist with the following. Eviews 10: I am estimating two linear equations that have the same explanatory variables, except for one variable that is defined differently for one equation versus another one. I need to test for statistical significance of the coefficients on this differently defined variable across two regressions. The possible solution I came across online is to implement such a test manually using the formula for two coefficients of interest:
his approach is to use the following Z test:
Z=(β1−β2)/square root of[(SEβ1)^2+(SEβ2)^2]
Where SEβ
is the standard error of β.
Is there any way to implement such a testing of equality of coefficients across models in Eviews 10 without manual derivations? Thank you in advance.
Please assist with the following. Eviews 10: I am estimating two linear equations that have the same explanatory variables, except for one variable that is defined differently for one equation versus another one. I need to test for statistical significance of the coefficients on this differently defined variable across two regressions. The possible solution I came across online is to implement such a test manually using the formula for two coefficients of interest:
his approach is to use the following Z test:
Z=(β1−β2)/square root of[(SEβ1)^2+(SEβ2)^2]
Where SEβ
is the standard error of β.
Is there any way to implement such a testing of equality of coefficients across models in Eviews 10 without manual derivations? Thank you in advance.