ARMAacf() will give you the acf for an autoregression, and toeplitz()
wil turn this into a correlation matrix. Then just multiply by the
desired variance.
I am not sure what this has to do with your subject line, and ?arima.sim
might be helpful for that.
On Fri, 17 Jun 2005, Caroline TRUNTZER wrote:
> Hello
> I would like to generate covariance matrix with autoregressive
> structure. I saw some functions in nlme such as corAR1 for example but I
> don't know how to use it for my goal. Could someone help me or advise
me
> another function?
> Thank you in advance
> Caroline
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595