[R] BackCast and Forecast with stats.arima and forecast.Arima

From: Paolo Rossi <statmailinglists_at_googlemail.com>
Date: Wed, 13 Apr 2011 14:14:45 +0100


Hello everyone,

I am experiencng some problems in producing forecasts and backcast from an ARIMA(1,0,0) model. I need to produce an insample backcast and a seasonal normal backcast and forecast.
I have a seasonal consumption function. By using actual data I get actual demand. By passing Seasonal Normal regressors I need to obtain seasonal normal demand.

I tested backcasting in arima by passing data used in the estimation back to the model through predict. This didnt work as the output was different from the fitted value I could see for the estimated model. In other words, in arima I am not able to produce my seasonal normal backcast.

I then started looking at Arima in Forecast. By passing the estimated model to Arima I can reproduce the fitted value from the model. My problem comes when I try to apply that model to different dependent (y) data, as the y is supposed to be passed to Arima, so essentially I can do only an in sample backcast for same dependent variable.

Is there a way I can get a Seasonal Normal backcast based on the model estimated on actual data but by passing Seasonal Normal regressors?

Thanks,

Paolo

        [[alternative HTML version deleted]]



R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Wed 13 Apr 2011 - 13:17:46 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Wed 13 Apr 2011 - 13:40:29 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive