0
$\begingroup$

Is it possible to use two dummy variables for breakpoints in a linear regression? In EViews I've created the following:

ls log(consumption) c log(gdp) log(gdp(-1)) log(consumption(-1)) @year>1989 @year>2008

The model is regressing household expenditure against GDP in Finland from 1970 to 2014 and the two dummy variables are an attempt to address breakpoints in the 2007/2008 financial crisis and the Finnish banking crisis of the 1990s.

$\endgroup$
0
$\begingroup$

breakpoints can either be in terms of level (i.e.intercepts ) or slopes . If you explicitely post your augmented data ( 3 columns ; n rows ) I will tell you which one your are testing.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.