Multiple graphs > boxplot

[dagrape]

Dear all

I am trying to represent a dependent variable (treatment) against different independent variables (v1, v2, v3....v20). I am using the following command:

boxplot(v1~treatment,data=y, main="xxxxxx",xlab="xxxxxx", ylab="xxxxxx")

However, it provides me only one graph for v1~treatment. For the other comparisons, I have to repeat the same command but changing the parameters. My intentions is to get different plots in just one sheet using only one command. Is it possible to join the same order for all the comparisons in only one command?

Thanks
David
     
        [[alternative HTML version deleted]]

______________________________________________
[hidden email] 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.

[David Winsemius]

On Oct 5, 2012, at 9:01 AM, David Gramaje wrote:

>
>
> Dear all
>
> I am trying to represent a dependent variable (treatment) against different independent variables (v1, v2, v3....v20). I am using the following command:
>
>
> However, it provides me only one graph for v1~treatment. For the other comparisons, I have to repeat the same command but changing the parameters. My intentions is to get different plots in just one sheet using only one command. Is it possible to join the same order for all the comparisons in only one command?
>

Two options:
A)
Melt the dataframe with reshape2::melt
then use boxplot( value ~ interaction(variable,treatment) )

B)
Melt the dataframe
Use bwplot in lattice with a slightly more full featured formula interface

bwplot(value~cat|variable, data=amelt)

Option B is more beautiful in my opinion.


--

David Winsemius, MD
Alameda, CA, USA


______________________________________________
[hidden email] 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.

[Rui Barradas]

Hello,

I've once written a function that does more or less what you want, but
it has no formula interface.

# Input:
#    x - matrix or data.frame of numeric vectors to graph
#    by - a factor or coercible to factor
multi.boxplot <- function(x, by, col=0, ...){
     x <- as.data.frame(x)
     uniq.by <- unique(by)
     by <- factor(by)
     len <- length(uniq.by) - 1
     n <- ncol(x)
     n1 <- n + 1
     col <- rep(col, n)[seq_len(n)]
     boxplot(x[[ 1 ]] ~ by, at = 0:len*n1 + 1,
         xlim = c(0, (len + 1)*n1), ylim = range(unlist(x)), xaxt = "n",
col=col[1], ...)
     for(i in seq_len(n)[-1])
         boxplot(x[[i]] ~ by, at = 0:len*n1 + i, xaxt = "n", add = TRUE,
col=col[i], ...)
     axis(1, at = 0:len*n1 + n1/2, labels = uniq.by, tick = TRUE)
}


a <- matrix(data=runif(300,max=2), nrow=100, ncol=3)
fac <- sample(letters[1:4], 100, TRUE)

multi.boxplot(a, fac)


Hope this helps,

Rui Barradas
Em 05-10-2012 17:01, David Gramaje escreveu:

>
> Dear all
>
> I am trying to represent a dependent variable (treatment) against different independent variables (v1, v2, v3....v20). I am using the following command:
>
> boxplot(v1~treatment,data=y, main="xxxxxx",xlab="xxxxxx", ylab="xxxxxx")
>
> However, it provides me only one graph for v1~treatment. For the other comparisons, I have to repeat the same command but changing the parameters. My intentions is to get different plots in just one sheet using only one command. Is it possible to join the same order for all the comparisons in only one command?
>
> Thanks
> David
>    
> [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] 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.

______________________________________________
[hidden email] 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.

[John Kane]

Does something like this make any sense?

library(reshape2)
library(ggplot2)
yy  <-  structure(list(A = c(23, 21, 21, 20, 19, 19), B = c(20, 18, 20,
19, 20, 18), C = c(15, 15, 15, 12, 13, 13)), .Names = c("A",
"B", "C"), class = "data.frame", row.names = c(NA, -6L))

y1  <-  melt(yy)  # using reshape2

ggplot(y1, aes(variable, value))+ geom_boxplot()

# or

ggplot(y1, aes(variable, value))+ geom_boxplot()  + facet_grid(variable ~ .)




John Kane
Kingston ON Canada

> -----Original Message-----
> From: [hidden email]
> Sent: Fri, 5 Oct 2012 18:01:39 +0200
> To: [hidden email]
> Subject: [R] Multiple graphs > boxplot
>
>
>
> Dear all
>
> I am trying to represent a dependent variable (treatment) against
> different independent variables (v1, v2, v3....v20). I am using the
> following command:
>
> boxplot(v1~treatment,data=y, main="xxxxxx",xlab="xxxxxx", ylab="xxxxxx")
>
> However, it provides me only one graph for v1~treatment. For the other
> comparisons, I have to repeat the same command but changing the
> parameters. My intentions is to get different plots in just one sheet
> using only one command. Is it possible to join the same order for all the
> comparisons in only one command?
>
> Thanks
> David
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] 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.

____________________________________________________________
FREE ONLINE PHOTOSHARING - Share your photos online with your friends and family!
Visit http://www.inbox.com/photosharing to find out more!

______________________________________________
[hidden email] 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.