[R] xyplot: Can I identify groups but not use them for regression?

Peter Ehlers ehlers at ucalgary.ca
Sat Sep 19 01:31:25 CEST 2009


And if you make 'year' a factor with levels '2007' and '2008'
you'll get your second wish.

Peter

Sundar Dorai-Raj wrote:
> I think this ought to work for you:
> 
> library(lattice)
> set.seed(42)
> d <- data.frame(year  = c(rep(2007,12), rep(2008,12)),
>                 treatment = rep(LETTERS[1:3], each = 4, times = 2))
> d$cover <- rnorm(nrow(d))
> d$variable <- rnorm(nrow(d))
> 
> xyplot(variable ~ cover | year, d,
>        panel = function(x, y, ...) {
>          panel.superpose(x, y, ...)
>          panel.lmline(x, y, ...)
>        },
>        groups = treatment)
> 
> HTH,
> 
> --sundar
> 
> On Fri, Sep 18, 2009 at 3:42 PM, Seth W Bigelow <sbigelow at fs.fed.us> wrote:
>> I wish to identify groups representing different treatments, but to plot
>> them and do a regression using a continuous variable ("cover")
>> ignoring the groupings.
>>
>> d$year <- NA
>> d$year <-c(rep(2007,12), rep(2008,12))
>> d$treatment <- c(rep("A",4),rep("B",4),rep("C",4), rep("A",4), rep("B",4),
>> rep("C",4))
>> d$cover <- rnorm(24)
>> d$variable <- rnorm(24)
>>
>> xyplot(variable ~ cover | year, d,
>>        type=c("p","r"),
>>        groups=treatment
>>        )
>>
>> As it stands, a different regression line is plotted for each treatment.
>> Oh, and how do I display the actual numeric value of year (e.g., "2007")
>> in the strip, rather than the word "year"?
>>
>> --Seth
>>
>>
>>
>> Dr. Seth  W. Bigelow
>> Biologist, USDA-FS Pacific Southwest Research Station
>> 1731 Research Park Drive, Davis California
>>        [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 
>




More information about the R-help mailing list