I'm trying to make the legend of this plot pretty, so I need there the be an actual superscript, which is why I am using the pretty10exp() function from the sfsmisc library. It works when I use the c() function.
However, I am also trying to keep the string and the scientific notation number on the same line. The legend() is broken into two lines, which I think is due to c(). I thought I could use paste(), but for some reason the output is now incorrect.
plot(1:12)
pVal <- 4
legend("topright", legend = c("P value:", sfsmisc::pretty10exp(pVal)), cex = 1.5)
legend("topright", legend = paste("P value:", sfsmisc::pretty10exp(pVal)), cex = 1.5)
pVal being an arbitrary number represented in scientific notation. The second line results in output like this: "P value: (significand) %*% 10^-4". The first line also doesn't give me what I want. How can I fix this problem?
pretty10exp returns an expression which allows it to use the ?plotmath features for making nice looking numbers. When working with expressions, you can't just paste values in like strings. You need to manipulate them with a special set of functions. One such function is substitute. You can do
plot(1:12)
pVal <- 4
legend("topright", cex = 1.5,
legend = substitute("P value: "*x, list(x=sfsmisc::pretty10exp(pVal)[[1]])) )
We use substitute() to take the value contained in the expression from pretty10exp and prefix it with the label you want. (We use * to concatenate rather than paste() since plotmath allows it)
This is what I would do:
fun <- function(text, pVal) {
y <- floor(log10(pVal))
x <- pVal / 10^y
bquote(.(text)*":" ~ .(x) %.% 10 ^ .(y))
}
plot.new()
text(0.5,0.7,fun("P value", 0.4))
text(0.5, 0.3, fun("P value", signif(1/pi, 1)))
No package is needed.
Related
I would like to add formatted text to a plot (title, axes labels, legend, etc.). A standard solution would be e.g.
plot(1, 1, ylab=substitute(lambda[1]))
where the y-axis label will be the greek letter λ with subscript 1. That is what I want but unfortunatelly it is hard-wired. I would like it to be flexible, i.e. to have an option to pass the text as an argument of a function. So, looking at r help, I found that the argument for substitute is an expression. So I've tried
sometext <- "lambda[1]"
e <- parse(text="sometext")
plot(1, 1, ylab=substitute(e))
But substitute ignores that e is an object an simply prints the text 'e' as the label. So instead I tried
plot(1, 1, ylab=eval(e))
which was better because eval at least evaluates the expression but the label now is literally 'lambda[1]', i.e. it is not evaluated as a greek letter λ with a subscript.
I then explicitly stated that 'e' is an expression doing
e <- expression(sometext)
and running the two previous plot commands but with the same results as before.
The closest I came to what I wanted ot achieve was doing
plot(1, 1, ylab=substitute(var[i], list(var="lambda", i=1)))
where at least the 1 is printed as a subscript but the text 'lambda' instead of the greek letter λ is printed.
Any suggestions on how can I achieve what the first command does but not hard-wired? E.g. using a string and somehow converting it to the right object so that it will be displayed properly? Thanks.
Daniel
You can pass the label as an expression like so and then use substitute inside the function:
dwat <- function(expr){
plot(1, 1, ylab = substitute(expr))
}
dwat(lambda[1])
dwat(mu[2])
If you want to pass a string instead of an expression use parse:
dwat_string <- function(string){
plot(1, 1, ylab = parse(text = string))
}
dwat_string("mu[1]")
Another option would be to use the ellipsis approach. That way you can pass an xlab or any other argument to plot:
dwat2 <- function(...){
plot(1, 1, ...)
}
dwat2(ylab = expression(lambda[1]))
you were close, checks if this works
substitute(lambda[i],list(i=i))
I'm trying to generate a beautiful legend in R plots. I have a factor=1e-5, that should appear nicely formatted in the legend. I found a nice function in the package sfsmisc, that transforms numbers to expressions. To add this expression to my bquote command, it seems that I need to transform itto a call. unfortunately, there are braces added at the end of the string (10^-5()).
Is there a way to avoid the addition of thoses braces? Or is there even an easier way to transform numbers to plotmaths commands for their use in legends? (without doing it manually)
factor = 1e-5
alpha = 1:10
omega = alpha^2 * factor
plot (
alpha
, omega
, xlab=bquote(alpha)
, ylab=bquote(omega)
, type="b"
)
text = expression()
# standard version
text[1] = as.expression(bquote(alpha%*%.(factor)))
# beautified version (use pretty10exp from sfsmisc package!?)
library("sfsmisc")
pretty = as.call(pretty10exp(factor, drop.1=T))
text[1] = as.expression(bquote(alpha^2%*%.(pretty)))
# add legend
legend("topleft", legend=text, pch=1, lty=1)
Here's what you can do instead with function parse:
text <- paste("alpha^2%*%",parse(text=pretty10exp(factor,drop.1=T)),sep="")
text
[1] "alpha^2%*%10^-5" # which we then use as the expression in your call to legend
legend("topleft", legend=parse(text=text), pch=1, lty=1)
See ?parse for more explanation on how this work.
Good day!
I am not that familiar to R so I'd be glad to get a little help.
Assume I have the following minimal example:
test <- c(10,20,40,80,80)
avg <- mean(test)
avg <- format(avg,digits=2)
plot(test, xlab="x", ylab="y", pch = 4)
legend("topleft", legend= c("Average: ", avg))
I'd like to write x̄ instead of "average" - wonder if this is event possible as it's not a regular symbol - merely a combination of two (letter plus overline).
The other thing I'd like to get rid of is the line break after the word "Average (see arrow in graphic below):
There are two issues here. The first is that this is handled using ?plotmath in R. The operator you are looking for is bar(). This is not a function but markup that plotmath understands.
The second is that you need an expression in which avg is converted to its value. You need an expression because that is what plotmath works with. There are several solutions to this problem, but the one I use below is bquote(). You provide it an expression and anything wrapped in .( ) will be converted its value by evaluating the thing inside the .( ).
Here is your code and a suitably modified legend() call:
test <- c(10,20,40,80,80)
avg <- mean(test)
avg <- format(avg,digits=2)
plot(test, xlab="x", ylab="y", pch = 4)
legend("topleft", legend = bquote(bar(x)*":" ~ .(avg)))
Do note that this will insert exactly what is in avg. You may need to do
avg <- round(avg)
or some other formatting fix to get something nice and presentable.
I refer to my previous question, and want to know more about characteristics of factor in R.
Let say I have a dataset like this:
temp <- data.frame(x=letters[1:5],
y=1:5)
plot(temp)
I can change the label of x easily to another character:
levels(temp[,"x"]) <- letters[6:10]
But if I want to change it into some expression
levels(temp[,"x"]) <- c(expression(x>=1),
expression(x>=2),
expression(x>=3),
expression(x>=4),
expression(x>=5))
The >= sign will not change accordingly in the plot. And I found that class(levels(temp[,"x"])) is character, but expression(x>=1) is not.
If I want to add some mathematical annotation as factor, what can I do?
I do not see any levels arguments in ggplot and assigning levels to a character vector should not work. If you are trying to assign expression vectors you should just use one expression call and separate the arguments by commas and you should use the labels argument in a scale function:
p <- qplot(1:10, 10:1)+ scale_y_continuous( breaks= 1:10,
labels=expression( x>= 1, x>=2, x>=3, x>= 4,x>=5,
x>= 6, x>=7, x>= 8,x>=9, x>= 10) )
p
I would just leave them as character strings
levels(temp[,"x"]) <- paste("x>=", 1:5, sep="")
If you then want to include them as axis labels, you could do something like the following to convert them to expressions:
lev.as.expr <- parse(text=levels(temp[,"x"]))
For your plot, you could then do:
plot(temp, xaxt="n")
axis(side=1, at=1:5, labels=lev.as.expr)
Expression is used to generate text for plots and output but can't be the variable names per se. You'd have to use the axis() command to generate your own labels. Because it can evaluate expressions you could try...
plot(temp, xaxt = 'n')
s <- paste('x>', 1:5, sep = '=')
axis(1, 1:5, parse(text = s))
Say, I have a variable rv which has some numerical value. Now, I want to plot the value of this variable on a base plot but preceded by a nicely formatted symbol e.g., r subscript m, using expression. To write on the plot I use mtext.
However, what I get is either the value of the variable, but no nicely formatted symbol (left annotation), or a nicely formatted symbol, but not the value of the variable, but the variable name...
I tried to play around with eval, but didn't get what I wanted. Here is my code:
plot(1:10, rep(10,10), ylim=c(0,12))
rv <- 0.43
#left annotation:
mtext(paste(expression(italic(r[M])), " = ", rv), side = 1, line = -1.5, adj = 0.1)
#right annotation:
mtext(expression(paste(italic(r[M]), " = ", rv)), side = 1, line = -1.5, adj = 0.9)
This is the result:
How do i get both, nice format and value of the variable? Thanks.
btw: I know that I can get it, if I use two times mtext and play around with adj and stuff. But I would really like to get it in one call or without playing around with the position of two annotations.
The bquote function will create an expression and alow substitution of values using .(var) syntax. for your case do something like:
text( 5,1, bquote( italic(r[M]) == .(rv) ) )
Just combine what you have and plot two pieces, joined by using adj:
R> plot(1:10, rep(10,10), ylim=c(0,12))
R> text(2,12, expression(paste(italic(r[M]))), adj=1)
R> text(2,12, paste("=", rv), adj=0)