[R-SIG-Finance] time series question

spencerg spencer.graves at prodsyse.com
Sat May 23 04:31:50 CEST 2009

Hi, Mark: 

      Have you considered using 'lme' in the 'nlme' package with a 
'corARMA' correlation structure, as described in sec. 5.3.3 of Pinheiro 
and Bates (2000) Mixed-Effects Models in S and S-PLUS (Springer)?  This 
package includes in a director " system.file('scripts', package='nlme')" 
files with names like "ch05.R" code to work essentially all the examples 
in the indicated chapters.  After you understand the contents of this 
book, and especially how to use this code for the type of problem you 
just described, you may wish to use the "dlm" package in conjunction 
with the "nlme" function in the "nlme" package.

      Hope this helps. 
p.s.  You mentioned "dlm".  In addition to having a vignette that helped 
me learn how to use it, it will soon have a companion book:  Petris, 
Petrone, and Campagnoli (2009) Dynamic Linear Models with R (Springer;  
scheduled to be available after June 26, according to Amazon).  I have 
not seen the book, but I like the "dlm" package, including the vignette. 

markleeds at verizon.net wrote:
> Hi everyone: Normally, if one has a single realization of a time series and one wants to estimate 
> say an ARMA(p,q) , where p and q are known ( for simplicity )  then one estimates it and that's that. 
> But, suppose that one has more than one realization  of the time series ( assuming each series is the same length) and yet still wants to estimate the "best" arma(p,q) , over all the realizations,  again where p and q are known. 
> I'm somewhat familiar with the literature but I don't know how to do this nor do I know of a book
> that talks about this problem.  The only thing I could think of was casting the arma(p,q) in its equivalent state space form and then possibly using the dlm package ?. Is this the only way to do this ? I was hoping that there was a simpler way ? or if anyone knows of a relevant paper or book, that would
> be appreciated.
> also, if assuming that p=1 and q=1 makes this question simpler, i'm willing to make that assumption also. thanks.
> _______________________________________________
> R-SIG-Finance at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only.
> -- If you want to post, subscribe first.

More information about the R-SIG-Finance mailing list