R help - Feb 2004 - How to predict ARMA models?

Hi all,
 
I am fitting an ARMA(1,(1,4)) model.
 
 y(t) = a*y(t-1) + e(t) + b1*e(t-1) + b4*e(t-4)
 
> arma1.14 <- arma(series, lag=list(ar=1, ma=c(1,4)), 
+  include.intercept = F, qr.tol = 1e-07) 
 
works fine:
 
  Coefficient(s):
       ar1     ma1     ma4  
     0.872  -0.445   0.331 
     
I want to forecast 50 periods.
I could not find a 'predict' function for ARMA models.
 
I can use predict() 
if I model an ARMA object as an ARIMA object:
i.e. ARIMA(1,0,1) for ARMA(1,1).
 
But I could not find a way to write 
 ARIMA(1,0,(1,4)) for ARMA(1,(1,4)).
 Using  arima(series,lag=list(ar=1, 0, ma=c(1,4)) , etc...  )
 does not work with ARIMA. 
 (Using 'seasonal = ...' won't do, either.)
 
Any clues would be greatly appreciated.
 
Thanks,
Joseph
 



---------------------------------

	[[alternative HTML version deleted]]