[R] ARIMA() fitting with XREG

Jan Verbesselt Jan.Verbesselt at agr.kuleuven.ac.be
Tue Jun 22 21:22:39 CEST 2004


Hi R-helpers,

I would like to derive a "realistic" R² between X(t)(time serie1) and
Y(t)(time serie2) from a fitted ARIMA model.

The actual ARIMA model is constructed like this;
Y[t] = a[1]Y[t-1] + ... + a[p]Y[t-p] + k*X(t)
 
*Y=Serie2
*X=Serie1=Xreg
*AR(p=8)

The correlation between X(t)(serie1) and Y(t)(serie2) is given by how a
large part of the variance is explained:  k*X(t)/a(p) (a(p)=> AR(8)).
If this is correct, how could I decompose the model to obtain this?

Thanks a lot!
Jan

> regressie1$arma
ar ma sar sma period diff sdiff
8  0   0   0     36    0     2


_______________________________________________________________________
Jan Verbesselt 
Research Associate 
Lab of Geomatics and Forest Engineering K.U. Leuven
Vital Decosterstraat 102. B-3000 Leuven Belgium 
Tel:+32-16-329750   Fax: +32-16-329760
http://gloveg.kuleuven.ac.be/




More information about the R-help mailing list