[R] Problem about SARMA model forcasting

[Saji Ren]

Somebody said that ARIMA models like discussed above are easy to implement on a spreadsheet.
The prediction formula is simply a linear equation that refers to past values of original time series and past values of the errors.
Thus, setting up an spreadsheet by stroing the data in one column, the forcasting values computed bu the formula in another column, and the errors(data minus forcasts) in the third column can give the results just as the same as any statistical computing software(like R).
But I just can't, and I think there must be some mistakes in my procedure but just can't find it by myself.
Really need helps!
Thanks again!

saji from Shanghai