On 18/01/2010 9:02 AM, Ivan Calandra wrote:> Hi everybody!
>
> I'm trying to write a script to plot a histogram, a boxplot and a
> qq-plot (under Windows XP, R2.10 if it matters)
>
> What I want to do: define the variables (x and y) to be used at the very
> beginning, so that I don't have to change all occurrences in the script
> when I want to plot a different variable.
>
> The dataset is called "ssfa". TO_POS is a categorical variable
> containing the tooth position for each sample. Asfc is a numerical
> variable. In my dataset, I have more variables but it wouldn't change;
I
> want to plot one numeric vs one category. Do I need to supply some data?
> I don't think it's really necessary but let me know if you would
like to.
>
> The code of what I do up to now:
> ---
> x <- ssfa$TO_POS
> y <- ssfa$Asfc
> hist(y, main="Histogram of Asfc", xlab="Asfc")
> boxplot(y~x, main="Boxplot of Asfc by TO_POS",
xlab="TO_POS", ylab="Asfc")
> ---
>
> I would like something like: hist(y, main="Histogram of y",
xlab="y")
> but that will add "Asfc" where I write "y".
> And the same for boxplot(y~x, main="Boxplot of y by x",
xlab="x", ylab="y")
> I thought about something like:
> ---
> cat <- "TO_POS"
> num <- "Asfc"
> x <- paste("ssfa$", "TO_POS", sep="")
> y <- paste("ssfa$", "Asfc", sep="")
> hist(y, main=paste("Histogram of ", cat, sep=""),
xlab=num)
> ---
> but it doesn't work since y is a string. I don't know how to get
the
> syntax correctly. I am on the right path at least?!
>
I think you're on the wrong path. You want to write a function, and
pass either x and y as arguments, or pass a formula containing both (the
former is easier). For example,
twoplots <- function(x, y) {
ylab <- deparse(substitute(y)) # get the expression passed as y
xlab <- deparse(substitute(x)) # get the expression passed as x
hist(y, main=paste("Histogram of ", ylab), xlab=ylab)
boxplot(y ~ x, main=paste("Boxplot of", ylab, "by",
xlab), xlab=xlab,
ylab=ylab)
}
Then
with(ssfa, twoplots(TO_POS, Asfc))
will give you your plots.
Duncan Murdoch