Page 1 of 1

forecasting with AR terms

Posted: Mon Nov 27, 2017 8:15 am
by russellrobins
i am sure there is a simple solution, but I can't seem to find it

i estimate an AR(1) model as in y c ar(1). I get the following results.

C -0.2113598143069835
AR(1) 0.7602108576310792
SIGMASQ 1.121348040513439

then I ask EVIEWS for the STATIC in-sample forecasts and I CAN't reproduce what EVIEWS produces as the in sample forecasts

shouldn't forecast(t) = -.2113 + .7602*y(t-1) reproduce the EVIEWS static in-sample forecasts? and yet, for each forecast I am off by a constant value

am i miss-interpreting the constant in this equation?

thanks

Russ

Re: forecasting with AR terms

Posted: Mon Nov 27, 2017 8:29 am
by EViews Gareth