Page 1 of 1

log deviation from trend (EViews)

Posted: Mon Jun 14, 2010 2:12 pm
by Dagoberd
Hi all,

Since I am new to econometrics and to EViews, I am struggling a bit...
I am dealing with a couple of time series variables.
First they should all be detrended with a linear and a quadratic trend. So I run the following regression (EViews):
x = c(1) + c(2)*@trend + c(3)*@trend^2
and save the residual series.

Then I need the log deviation from trend for one variable, and the level deviation from trend for another, but I cannot figure out how to do this...
Can anyone shed some light on this? It would be greatly appreciated!

Thx,
Dieter