Good afternoon!
I'm trying to learn time series but i have a bit of of a problem using R
packages for this.
1.
> LakeHuron
> sample(500:600, 98)
> sample(500:600, 98)->t
> fit<-arima(LakeHuron, order=c(2,1,1), xreg=t)
> fit
> predict(fit, n.ahead=1, newxreg=t)
Now, my problem is this: is it ok to use the same t in predict function or
should my newxreg contain 99 values in order to properly predict the 99th value
for LakeHuron?
Another problem is that i don't receive the 99th value on the screen, as i
should (running the example in ?predict
> (fit <- arima(USAccDeaths, order = c(0,1,1),
+ seasonal = list(order=c(0,1,1))))
> predict(fit, n.ahead = 6)
gives me on the screen the predicted values, so i think I must be doing
something wrong)
2. If i have ts for 3 regions of the country, is it ok to model each one
independent of each other ? or is there a way of modeling the whole ts but
keeping in the analysis a dummy variable for the region (thou this seems
unrealistic)
3. In the dyn package does my ts have to fulfill the same properties of
stationary ts? Is it ok to regress a dependent time series on some other time
series, as dyn does? How can one predict a future value of a ts using dyn
package, for example how can i predict the 99th value for the LakeHuron data
using dyn (the pdf ofr this pack is unclear to me)?
Thank you and have a great day!
---------------------------------
Be a better sports nut! Let your teams follow you with Yahoo Mobile. Try it now.
[[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.
