No you cannot, strictly speaking. You could manually enforce the
shortest of those samples for all specifications and then make your choice.
Am 25.07.2011 10:44, schrieb Sam Sam:
If I use the monthly data from 1980M1 to 1990M12(the number of samples
is 132)，and there are there ARIMA model: ARIMA(1,1,1), ARIMA(2,1,0),
and ARIMA(1,1,0)(1,1,0)_12 . Can I choose the model by AIC or BIC from
these three models when the number of using observations different?
When estimating the model,
The using observations of ARIMA(1,1,1) : 1980M3 to 1990M12. (number:130)
The using observations of ARIMA(2,1,0):1980M4 to 1990M12. (number:129)
The using observations of ARIMA(1,1,0)(1,1,0)_12 :1983M3 to 1990M12.
Can I compare AIC of these three models when the number of using
observations are different ?
How should I choose the model by AIC? Is the basis of comparison different?
Gretl-users mailing list