[R] Help with ARIMA seasonal time series

rkevinburton at charter.net rkevinburton at charter.net
Sat Sep 13 08:37:36 CEST 2008

I have a set of data that is basically 3+ years of data. It is daily sales for this year and then back 3 years so there are 3*365 + 231 days or 1326 days of data. Since this is a time series I have constructed it as:

Start = c(1, 1) 
End = c(4, 231) 
Frequency = 365 

In trying to ayalyze this data I first turned to stl to break the time series into components. This worked really well but I would like to be able to forecast and my attempts so far at applying ARIMA to the "remainder" portion of the stl output have failed. So I decided to try a different approach. I decided to let arima handle the seasonal part as well. So I tried:

tarima = arima(TimeSeries, order=c(1,1,1), seasonal=list(order=c(1,1,1)))

But this call results in the error:

Error in makeARIMA(trarma[[1]], trarma[[2]], Delta, kappa) : 
  maximum supported lag is 350

Is it not possible to use ARIMA to fiit daily seasonsal data as I have described? Am I missing something fundamental?

Thank you.


More information about the R-help mailing list