Re: [R] Fwd: time series patterns

From: Spencer Graves <>
Date: Sun 09 Jul 2006 - 04:50:38 EST

          For my previous email on this thread, I couldn't find Jim Lindsey's home page. I just found it via -> Search -> "R Site Search" -> "Jim Lindsey's packages." This led me to "".

	  ... just in case someone following this thread also has this problem.
	  Spencer Graves

	  I how you code something like this depends on what you plan to do
with it. If it were my problem, I might want to try some hidden Markov models. RSiteSearch("hidden markov", "functions") produced 29 hits for me just now. You might try some of the packages mentioned there. They might suggest supplemental coding and / or plotting procedures that might provide insight into your application.
	  Hope this helps.
	  Spencer Graves

p.s. The 'repeated' package and others by J.K. Lindsey are not available from CRAN. A year ago, I downloaded material from his personal web site, which was the first hit on a Google search for "J K Lindsey". I also remember seeing it on CRAN or Today, the closest I can come is "" (TÚl : +32 4 3662964). If you want to try any of his R packages and you don't have any better results with a web search than I got just now, I suggest send him an email.

Raphael Fraser wrote:
> ---------- Forwarded message ----------
> From: Alexander Nervedi <>
> Date: Jun 30, 2006 4:56 PM
> Subject: time series patterns
> To:
> Hi all.
> I have a factor variable distributed over time. I am looking for an elegant
> way to code duration of a state. Suppose,

>> rainfall.shocks <- factor(sample(c(1,2,3), size = 15, replace = TRUE, prob
>> = unit.p),

> + label = c("Drought", "Normal", "High"))
>> rainfall.shocks

> [1] Normal High High Drought Normal Normal High Normal Drought
> [10] Normal Drought Normal Normal Normal Normal
> So capture the duration of say drought, I'd need a variable that is able to
> keep track of rainfall.shocks as well as its past values. I was wondering if
> there is any obvious way to do this. the Drought variable in this case would
> have values
> 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0
> many thanks for the suggestions you are likely to make.
> Alexander Nervedi
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today - it's FREE!
> ______________________________________________
> mailing list
> PLEASE do read the posting guide! mailing list PLEASE do read the posting guide! Received on Sun Jul 09 04:54:34 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Sun 09 Jul 2006 - 06:16:26 EST.

Mailing list information is available at Please read the posting guide before posting to the list.