[R] problems whit seasonal ARIMA

[Rolf Turner]

On 2/11/2009, at 2:49 AM, Laura Saltyte wrote:

> Hello,
>
> I have daily wind speed data and need

	``need'' is probably not the operative word!

> to fit seasonal ARIMA model, problem
> is that my period is 365. But when I use arima(...) function, with  
> period
> 365, I’m getting error message: “Error in makeARIMA(trarma[[1]],
> trarma[[2]], Delta, kappa) :   maximum supported lag is 350”. Can  
> someone
> help me with this problem?

	Well, you could dig into the code and rewrite it to revise the maximum
	lag.  But that would be pretty silly.

	A seasonal model with a lag of 365 is downright ridiculous.

	A seasonal model of lag ``s'' assumes that X_{t-s} has an influence
	upon or contains information about (or is non-trivially correlated
	with) X_t.  Do you *really* believe that wind speed from a year ago
	contains information relevant to current wind speed?

	There is almost surely a periodic *trend* with period 365 (well, maybe
	period 365.25?) in daily wind speed data, but that's a different issue.
	If you have data for sufficiently many years you could try estimating
	such a trend simply by averaging over all occurrences of a given day to
	estimate the trend value for that day.  You could also try fitting a
	sinusoid (probably a fairly high order sinusoid ....)

	What you really should do depends on the (practical) question that
	you are trying to answer.

		cheers,

			Rolf Turner
######################################################################
Attention: 
This e-mail message is privileged and confidential. If you are not the 
intended recipient please delete the message and notify the sender. 
Any views or opinions presented are solely those of the author.

This e-mail has been scanned and cleared by MailMarshal 
www.marshalsoftware.com
######################################################################