Page 1 of 1

Structural Breaks in a Series

Posted: Sun Nov 14, 2010 10:11 am
by Hurlstrino
Hi,

I know that there are questions about testing for structural breaks once you have an equation.

My question is how I can test a series I already have for structural breaks. Eviews always wants me to use the equation, however it is the series I am interested in, not the equation.

Can this be done ?

Thanks

Re: Structural Breaks in a Series

Posted: Sun Nov 14, 2010 11:57 am
by EViews Gareth
Can you define what a structural break is "in a series"?

Re: Structural Breaks in a Series

Posted: Sun Nov 14, 2010 12:04 pm
by Hurlstrino
OK,

So I am trying to run a granger causality analysis on 20 series, each containing quarterly data.

It is my understanding that before I run the granger analysis, I need to take into account the structural breaks.

Now, due to the fact that I am running the causality analysis on a group of 20 series, and I am trying to see if

Series 1 -> Series 2
Series 2 -> Series 3
Series 3 -> Series 1 etc

I don't have just two series that I can create an equation out of.

1. How would I go about testing for the structural breaks in my data ?
2. Do I need to worry about the structural breaks to be able to run the granger analysis, or are the breaks irrelevant ?


Thank you so much for your prompt reply !