R help - Apr 2007 - Questions about statistics: ARIMA modeling seasonal MA term close to -1

Hello,

I am trying to fit a seasonal ARIMA model. I am using the ARIMA package in
R. when I look at the model coefficients I find that that the first order
seasonal moving average term is very close to -1(this is after first order
seasonal differencing), I have tried to look at the literature on unit roots
etc..., however I have not been able to find a good explanation of what the
implications are for forecasting when using a model with a MA term close to
-1. Does anyone have any suggestions how to deal with the described
situation?

Best Regards,

Spencer Jones

	[[alternative HTML version deleted]]