R help - Feb 2009 - lattice shingle with time and date format

Hi R-Users,

I have a time series of bivariate observations (x,y,t).  I've  
developed a few panel routines to explore certain aspects of the data,  
and now would like to break the time series up into smaller chunks to  
explore the temporal trends.  I would like to know if anyone has any  
experience breaking up time series with lattice.

Base graphics offers coplot, utilizing the co.intervals routine to  
break up the series into equal chunks.  Lattice has equal.count as a  
wrapper for co.intervals.  As you can see in the example below, I can  
recreate a 'coplot' with lattice but am having difficulty getting the  
x-axis of the shingle plot and the strip labeling on the xyplot to be  
in time or date format.  My question really consists of two parts,  
with the second part being less important for my data visualization  
purposes:

1) How might I be able to convert the x-axis of the shingle plot from  
numeric to date format?
2) How might I be able to customize the strip labels on the xyplot to  
show a range, like min(month, year)-max(month, year) for each panel?

Thank you very much for any helpful suggestions.

Jon
#########################################
# Example R script

library(lattice)

times<-as.Date(seq(11400,14190,
length.out=100),origin="1970-01-01")
x<-rnorm(100)
y<-rnorm(100)
df<-data.frame(x=x,y=y,times=times)

#I want to recreate something like coplot with lattice
coplot(y~x|times,data=df)


#this doesn't work, which is an issue with co.intervals
breaks<-equal.count(times)

#this does work
breaks<-equal.count(unclass(times))


#In the lattice book there is this outline
bPlot<-plot(breaks)
dPlot<-xyplot(y~x|breaks, data=df)
plot(bPlot, position=c(0,0.65,1,1))
plot(dPlot, position=c(0,0,1,0.65),newpage=FALSE)
#The plot is ok, but the xaxis is in numeric and not date
#########################################


Jon Loehrke
Graduate Research Assistant
Department of Fisheries Oceanography
School for Marine Science and Technology
University of Massachusetts
200 Mill Road, Suite 325
Fairhaven, MA 02719
jloehrke@umassd.edu

 > sessionInfo()
R version 2.8.1 (2008-12-22)
i386-apple-darwin8.11.1

locale:
en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] grid      stats     graphics  grDevices utils     datasets   
methods   base

other attached packages:
[1] lattice_0.17-17
	[[alternative HTML version deleted]]

On Tue, Feb 10, 2009 at 6:29 AM, Jon Loehrke <jloehrke at umassd.edu>
wrote:
> Hi R-Users,
>
> I have a time series of bivariate observations (x,y,t).  I've
> developed a few panel routines to explore certain aspects of the data,
> and now would like to break the time series up into smaller chunks to
> explore the temporal trends.  I would like to know if anyone has any
> experience breaking up time series with lattice.
>
> Base graphics offers coplot, utilizing the co.intervals routine to
> break up the series into equal chunks.  Lattice has equal.count as a
> wrapper for co.intervals.  As you can see in the example below, I can
> recreate a 'coplot' with lattice but am having difficulty getting
the
> x-axis of the shingle plot and the strip labeling on the xyplot to be
> in time or date format.  My question really consists of two parts,
> with the second part being less important for my data visualization
> purposes:
>
> 1) How might I be able to convert the x-axis of the shingle plot from
> numeric to date format?
Perhaps something like

plot(breaks, xlim = as.Date(extendrange(breaks), origin = structure(0,
class = "Date")))

or the simpler

plot(breaks, xlim = extendrange(times))

(the warning seems harmless).
> 2) How might I be able to customize the strip labels on the xyplot to
> show a range, like min(month, year)-max(month, year) for each panel?
br.levels <- sapply(levels(breaks),
                    function(x) {
                        paste(as.character(structure(x, class =
"Date")),
                              collapse = ", "))

xyplot(y~x|breaks, data=df,
       strip = strip.custom(factor.levels = br.levels,
                            strip.levels = TRUE,
                            strip.names = FALSE))

You probably want a suitable 'format' in the as.character() call.

-Deepayan