Re: [Gretl-users] Estimation of ARIMA
Sam,
Q1:
I input monthly data from 1980M1 to 1990M12. I set the
sample range from 1980M6 to 1990M12. When I estimate AR(3) model, data used to estimate
the model is 1980M1-1990M12, 1980M6-1980M12, 1980M9-1990M12, or so on?
What I mean is when estimating parameter of AR(3) model or residual of AR(3) model, is
data from 1980M1 to 1980M6 included for estimating?
Yes. gretl uses pre-sample values for lags when they're available.
Q2:
I input monthly data from 1980M1 to 1990M12. I set the
sample range from 1980M1 to 1988M12(I want to use data from 1989M1 to 1990M12 for out-of
sample forecasting)
<snip>
Can I compare MAPE(Mean Absolute Percentage Error) value (from 1989M1
to 1990M12) of these three models when their period and number of using observations are
different?
Yes. What matters is that the forecast range is common to all models. They don't have
to be estimated over a common sample.
PS