Structural Break in Return Series Please Help!
Posted: Tue Oct 15, 2013 3:39 am
Hey guys,
This is my first post on the forum. I have been learning to use Eviews recently (version 7) and am stuck on a problem I have been given.
I have been given data on the weekly returns of the Dow Jones Composite Average (DJCA) index. The data ranges from 1954-01-13 to 2012-06-27. At the 821st observation, I have been told that there is a structural break in volatility.
What I want to do is test whether or not this structural break is statistically significant. I am using weekly squared returns as a proxy for volatility.
My initial thought has been to run a regression of squared returns on a constant and include a dummy variable for the structural break. I was then going to test the significance of the coefficient on the dummy variable.
Can anyone think of a problem with this or a better way of doing it. I know people use the Chow Test or CUSUM for testing structural breaks. I am just not sure how to use those tests in the context of my return series.
Any help is appreciated.
cheers
This is my first post on the forum. I have been learning to use Eviews recently (version 7) and am stuck on a problem I have been given.
I have been given data on the weekly returns of the Dow Jones Composite Average (DJCA) index. The data ranges from 1954-01-13 to 2012-06-27. At the 821st observation, I have been told that there is a structural break in volatility.
What I want to do is test whether or not this structural break is statistically significant. I am using weekly squared returns as a proxy for volatility.
My initial thought has been to run a regression of squared returns on a constant and include a dummy variable for the structural break. I was then going to test the significance of the coefficient on the dummy variable.
Can anyone think of a problem with this or a better way of doing it. I know people use the Chow Test or CUSUM for testing structural breaks. I am just not sure how to use those tests in the context of my return series.
Any help is appreciated.
cheers