Can you clarify what exactly you mean by this? "[N]ow [I] would like to use the last X values to predict tomorrow's weather. I'm at a loss. All the functions I've come across (like forecast()) use the series and then forecast from the end point."
It sounds like a prediction to me. Anyways, I think most methods do allow "new" values for the independent variables: e.g., the newdata argument to most predict() methods and the xreg arguments to forecast::forecast(). Do you know which method you are using? Hope this helps, Michael On Wed, Jan 18, 2012 at 4:17 PM, nhomeier <nhome...@aer.com> wrote: > Couldn't find this in the archives. I'm fitting a series of historical > weather-related data, but would like to use the latest values to forecast. > So let's say that I'm using 1970-2000 to fit a model (using fourier terms > and arima/auto.arima), but now would like to use the last X values to > predict tomorrow's weather. I'm at a loss. All the functions I've come > across (like forecast()) use the series and then forecast from the end > point. > > Do I need to decompose the fit and write it out the long way? For example, > Tomorrow = fit$coef[1]*Yesterday + fit$coef[2]*BeforeYesterday + etc > > or is there a function that I'm not finding? > > Thank you, > Nicole > > -- > View this message in context: > http://r.789695.n4.nabble.com/forecasting-a-time-series-tp4308147p4308147.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > 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. ______________________________________________ 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.
