[R] Functions for autoregressive Regressionmodels (Mix between times series and Regression Models) ?

Achim Zeileis Achim.Zeileis at wu-wien.ac.at
Tue Aug 7 16:01:44 CEST 2007

On Tue, 7 Aug 2007, Gabor Grothendieck wrote:

> arima (see ?arima) supports explanatory variables in the xreg= argument.

In addition to arima(), the function dynlm() in package "dynlm" might be
useful if you want to fit the model by OLS. See also package "dyn" for a
similar approach.
Z

> On 8/7/07, "Maja Schröter" <maja.schroeter at gmx.de> wrote:
> > Hello everybody,
> >
> > I've a question about "autoregressive Regressionmodels".
> >
> > Let Y[1],.....,Y[n], be a time series.
> >
> > Given the model:
> >
> > Y[t] = phi[1]*Y[t-1] + phi[2]*Y[t-1] + ... + phi[p]*Y[t-p] + x_t^T*beta + u_t,
> >
> >
> > where x_t=(x[1t],x[2t],....x[mt]) and beta=(beta[1],...,beta[m]) and u_t~(0,1)
> >
> > I want to estimate the coefficients phi and beta.
> >
> > Are in R any functions or packages for "autoregressive Regressionmodel" with special summaries?. I'm not meaning the function "ar".
> >
> >
> > Example: I have the data
> >
> > working.time        <- rnorm(100)   # Y
> > vacation            <- rnorm(100)   #x1
> > bank.holidays       <- rnorm(100)   #x2
> > illnes              <- rnorm(100)   #x3
> > education           <- rnorm(100)   #x3
> >
> >
> >
> > Now I want to analyse:
> >
> >  Y[t] = phi[1]*Y[t-1] + phi[2]*Y[t-1] + ... + phi[p]Y[t-p] + beta1*vacation_t  +beta2*bank.holidays + beta3*illnes + beta4*eductation + u_t-
> >
> >
> >
> > Has anyone an idea?
> >
> > I would be more than glad if so.
> >
> > Thank you VERY much in advance.
> >
> > Kindly regards from the Eastern Part of Berlin,
> >
> > Maja
> >
> > --
> >
> > ______________________________________________
> > R-help at stat.math.ethz.ch mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > and provide commented, minimal, self-contained, reproducible code.
> >
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help