[R] Fwd: time series patterns

Spencer Graves spencer.graves at pdf.com
Wed Jul 5 03:17:25 CEST 2006


	  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 r-project.org.  Today, 
the closest I can come is "jlindsey at ulg.ac.be" (Tél : 	+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 <alexnerdy at hotmail.com>
> Date: Jun 30, 2006 4:56 PM
> Subject: time series patterns
> To: raphael.fraser at gmail.com
> 
> 
> 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!
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
> 
> ______________________________________________
> 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



More information about the R-help mailing list