Page 1 of 1

Chow and RESET

Posted: Mon Jan 12, 2015 5:57 pm
by Nuttela
I have an urgent question, would really appreciate someone's help.

I have an OLS cross-sectional model with dummy variables (attached in the file). The data set contains two years 2011/2013.

1) In model 7, should the year be treated as a constant in the model? Do we just exclude it from independent variables in this case? I have transformed the years into dummies with 2013=1 and 2010. However, when performing the Chow test, I keep receiving an error of When the year is removed, the error disappears. Could someone explain...?

2) Also, when dummies are included the RESET test suggests non-linearity (Model 7). However, the functional form is linear in Model 2 where there are no dummies. Is it normal for dummies to affect the linearity? Any suggestion why this happens?

Thank you for your help.

Best,
Nataly