R help - Mar 2007 - Heteroskedastic Time Series

Hi R-helpers,

I'm new to time series modelling, but my requirement seems to fall just
outside the capabilities of the arima function in R.  I'd like to fit an
ARMA model where the variance of the disturbances is a function of some
exogenous variable.  So something like:

Y_t = a_0 + a_1 * Y_(t-1) +...+ a_p * Y_(t-p) + b_1 * e_(t-1) +...+ b_q *
e_(t-q) + e_t,

where

e_t ~ N(0, sigma^2_t),

and with the variance specified by something like

sigma^2_t = exp(beta_t * X_t),

where X_t is my exogenous variable.  I would be very grateful if somebody
could point me in the direction of a library that could fit this (or a
similar) model.

Thanks,

James Kirkby
Actuarial Maths and Stats
Heriot Watt University

check fseris library.

On 3/5/07, james at ma.hw.ac.uk <james at ma.hw.ac.uk>
wrote:
> Hi R-helpers,
>
> I'm new to time series modelling, but my requirement seems to fall just
> outside the capabilities of the arima function in R.  I'd like to fit
an
> ARMA model where the variance of the disturbances is a function of some
> exogenous variable.  So something like:
>
> Y_t = a_0 + a_1 * Y_(t-1) +...+ a_p * Y_(t-p) + b_1 * e_(t-1) +...+ b_q *
> e_(t-q) + e_t,
>
> where
>
> e_t ~ N(0, sigma^2_t),
>
> and with the variance specified by something like
>
> sigma^2_t = exp(beta_t * X_t),
>
> where X_t is my exogenous variable.  I would be very grateful if somebody
> could point me in the direction of a library that could fit this (or a
> similar) model.
>
> Thanks,
>
> James Kirkby
> Actuarial Maths and Stats
> Heriot Watt University
>
> ______________________________________________
> R-help at stat.math.ethz.ch 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.
>

-- 
WenSui Liu
A lousy statistician who happens to know a little programming
(http://spaces.msn.com/statcompute/blog)