Hi Michael, As far as I know, there is no Arima function in R that allows multiple seasonal patterns. There's 2 approaches you can take: 1. Use a matrix of dummy variables (e.g. jan/feb/mar/...dec and mon/tues/wed/...sun) as an xreg term in the arima function. 2. Use a matrix of Fourrier terms to represent the different seasonal patterns.
Rob Hyndman outlines the second approach on his blog<http://robjhyndman.com/researchtips/longseasonality/>, which might be a good starting point. The primary disadvantage of these approaches is that the seasonal term is assumed to be fixed over time. -Zach On Sun, Dec 11, 2011 at 9:56 PM, Michael <comtech....@gmail.com> wrote: > sorry for the re-post... > > On Sun, Dec 11, 2011 at 8:22 PM, Michael <comtech....@gmail.com> wrote: > > > Hi all, > > > > I just couldn't find a R function which can fit multiple seasonal > > patters... i.e. in the following code: > > > > *arima(x = data, order = c(p, d, q), seasonal = list(order = c(P, D, Q), > > period = S), ... > > *** > > * > > there can be only one "period", am I right? > > > > What if the data seem to have three different seasonality cycles, 5, 12, > > 21? > > > > Thanks a lot! > > * > > > > [[alternative HTML version deleted]] > > _______________________________________________ > r-sig-fina...@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-sig-finance > -- Subscriber-posting only. If you want to post, subscribe first. > -- Also note that this is not the r-help list where general R questions > should go. > [[alternative HTML version deleted]] ______________________________________________ R-help@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.
