Time Series Checks and Structural Break
Posted: Tue Jan 05, 2016 8:31 am
Hi All,
I am running a time series model of how average weight varies across time for different countries. I want to see if there is a structural change after a particular year and I control for various aggregate indicators (per capita income, number of fast food restaurants, activity levels etc) :
W(t) = a + Y(t) + R(t) + A(t) + error
I had two questions firstly when estimating this particular model. Firstly are there any checks I should carry out before estimating this model (e.g. cointegration, unit roots etc.) I expect these variables to have a causal effect, but my main aim is to see if after a particular year having controlled for these variables whether there is a jump in the growth rate of weight. So is this model appropriate. In addition if the beginnings of the model are appropriate would a Chow test suffice for testing the structural break.
I would really appreciate any help!
p.s. I'm using Eviews 8.1
I am running a time series model of how average weight varies across time for different countries. I want to see if there is a structural change after a particular year and I control for various aggregate indicators (per capita income, number of fast food restaurants, activity levels etc) :
W(t) = a + Y(t) + R(t) + A(t) + error
I had two questions firstly when estimating this particular model. Firstly are there any checks I should carry out before estimating this model (e.g. cointegration, unit roots etc.) I expect these variables to have a causal effect, but my main aim is to see if after a particular year having controlled for these variables whether there is a jump in the growth rate of weight. So is this model appropriate. In addition if the beginnings of the model are appropriate would a Chow test suffice for testing the structural break.
I would really appreciate any help!
p.s. I'm using Eviews 8.1