To decompose dependent variable from a regression

[micallefb]

Hi,

I am currently doing an analysis of non-performing loans in my country using an error-correction approach and would like to decompose the (year-on-year growth) of the dependent variable (non-performing loans) into the contribution of the explanatory variables (e.g. income, house prices, interest rates etc), in addition to the standard analysis like interpretation of regression results and simulation.

Is there an easy way to do it in Eviews? Or have someone already written a programme to perform this decomposition?

Thanks

B.

[jfgeli]

I am also trying to do this. I understand that you want to be able to decompose the growth rate of your explained variable according to the evolution of the explanatory ones, right?
So far the only approach I see feasible is the following:
1) Collect the coefficients from the equation representation output.
2) Rewrite the estimated equation so as to have your explanatory variables appearing only once on the right hand side.
3) Create series of contributions using the coefficients resulting from 1) and 2) and each explanatory variable.

Dear Eviews´s Gareth, Glenn and company, is there a way to do this easily on Eviews. At least to rewrite the equation so we can programmatically do the rest?

Cheers,

Fede