Re: [R] auto.arima help

2008-09-26 Thread rkevinburton
Yes, sorry it auto.arima is in the 'forecast' package. The following produces the problem: auto.arima(ts(c(rep(0,104), rep(143, 52), rep(260,33)), frequency=52)) Kevin Prof Brian Ripley <[EMAIL PROTECTED]> wrote: > On Thu, 25 Sep 2008, [EMAIL PROTECTED] wrote: > > > I am calling auto.ar

Re: [R] auto.arima help

2008-09-26 Thread Prof Brian Ripley
On Thu, 25 Sep 2008, [EMAIL PROTECTED] wrote: I am calling auto.arima with a time series that is about 186 observations long with a frequency of 52. With some time series I get: 1:last.nonzero: result would be too long a vector Is there something that I can do to the data to avoid this error?

[R] auto.arima help

2008-09-25 Thread rkevinburton
I am calling auto.arima with a time series that is about 186 observations long with a frequency of 52. With some time series I get: 1:last.nonzero: result would be too long a vector Is there something that I can do to the data to avoid this error? Thank you. Kevin

[R] auto.arima help.

2008-09-22 Thread rkevinburton
Hello, I am calling the auto.arima method in the forecast package at it returns what seems to be valid Arima output. But when I feed this output to 'predict' I get: Error in predict.Arima(catall.fit[[.index]], n.ahead = 12) : 'xreg' and 'newxreg' have different numbers of columns Is there a