I make some plots with R and use bquote because I need variables for the main of the plot. However, the main is no longer bold but I want it to be bold. I defined the main as follows:
title = bquote(atop("Empirical Pricing Kernel at Date",~.(EndDate)~"with Index Price"~.(ST)~"€"))
plot(temp, EPK, type="l", main = title)
Enddate contains "2014-08-01" as date and ST is just numeric with 9210.08.
Is there any way to make it bold with or without bquote? I'd like to find a solution with bquote because it's very convenient when using subscripts.
My problem is that I am using it in a par-plot with two plots and the other plot needs no special things in it's main. So, the main is bold. I even tried to just put bquote around it in order to get the same font size but it stayed bold.
I prefer to use what I think of as "pure plotmath" so I use tilde's instead of spaces and use no quotes. I suspect it was the leading tilde in the second argument to bquote that was throwing the error. In plotmath the tildes need something on either side: If you really need a none-displayed something you can always use phantom(0) but why bother in this case?
bquote(atop(Empirical~Pricing~Kernel~at~Date,
bold(.(EndDate))~with~Index~Price~.(ST)~"€"
) )
Test:
EndDate="2014-08-01";ST=9210.08
title = bquote(atop(Empirical~Pricing~Kernel~at~Date, bold(.(EndDate))~with~Index~Price~.(ST)~"€"))
plot(1,1, type="l", main = title)
Related
I'm trying to create a list of labels that contain italics. I can do it with "expression" like this, and when I put them on a plot (by adding a legend as an example, but I'll use them different ways), it all works nicely.
sp.names=c(expression(paste("species ",italic("one")," sp.")),
expression(paste("species ",italic("two")," sp.")))
plot(1:10)
legend("topleft",legend=sp.names)
But instead of specifying the words in the label directly in the code, I want to call them from cells in a dataframe (because there are a lot of them and they change depending on my underlying data). But when I try and specify which dataframe cell I want, it doesn't print the labels correctly (see below). Perhaps there is a different way for me to call the cell that I want that the "expression" function will recognise?
df=data.frame(V1=c("species","species","species"),V2=c("one","two","three"))
sp.names=c(expression(paste(df$V1[1],italic(df$V2[1])," sp.")),
expression(paste(df$V1[2],italic(df$V2[2])," sp.")))
plot(1:10)
legend("topleft",legend=sp.names)
Use substitute, it substitutes variables in expressions.
sp.names=c(substitute(V1 ~ italic(V2) ~ "sp.", df[1,]),
substitute(V1 ~ italic(V2) ~ "sp.", df[2,]))
I also removed the unneeded paste (which has a different meaning within plotmath) and replaced it with ~ for increased readability.
Give bquote a shot.
sp.names <- c(bquote(.(df$V1[1])~italic(.(df$V2[1]))~" sp."),
bquote(.(df$V1[2])~italic(.(df$V2[2]))~" sp."))
plot(1:10)
legend("topleft", legend=sp.names)
Specifying text in a base R plot() with formatting such as italics / bold font / newline usually involves one or more of the following functions:
paste()
expression()
atop()
substitute()
italic()
Is there an intuitive explanation for the differences between these functions and when best to apply them?
What you're referring to is the plotmath syntax.
To start off, let's make it clear that for a plotmath expression to be interpreted as such, you tell R it's an "expression" and that is why you need expression().
So any time you want to use special symbols or formatting, like italic() and atop(), it's actually a part of plotmath and so you need to wrap it in an expression. eg:
plot(0, main = expression(atop(over,italic(under))))
If you've tried out ?italic or ?atop, you've probably noticed it takes you straight to the plotmath manual page, where a bunch of other functions are listed.
What about substitute() ? Well in my previous example, you'll notice I used strings directly to write 'over' and 'under', without putting them within quotes. This is because of the special expression() environment.
So if you need to put whatever is inside a variable in your text (rather than the variable name) then you put your expression inside a substitute() and give it the arguments. eg:
plot(0, main = substitute(atop(oo,italic(under))), list(oo='over2')))
Note that we don't put substitute around the expression block but replace it entirely.
Finally, where does paste() come in all this ? Well, paste is the glue (pun intended) with any text not dealt with by plotmath.
So if you need text before or after math symbols (or formatted text), you paste() things together within the expression (or substitute) environment. eg :
plot(0, main = substitute(paste("b4", atop(oo,italic(under)), aft),
list(oo='over', aft = 'after3')))
As before, if you want to paste the content of a variable, you need substitute.
And Voilà that's most of the plotmath you'll ever need!
For any other symbols, or functions, have look at ?plotmath
I've learned to apply bold to a portion of the text used in a plot title using mtext() expression() and paste(). This works great if you specify the strings outright. However, in the project I'm working on now, the portion of text to be bolded needs to be obtained through a call to an element of a vector. However, the characters needed in the call syntax are interpreted by expression() and the call fails.
junk <- c("I'm Special", "You're Special")
plot(0, type="n")
mtext(expression(paste("Do you think ", bold(junk[1]),"today?")),3,2)
mtext(expression(paste("I think ", bold(junk[2]), "today.")),3,1)
Any thoughts on how to approach this? I am trying to avoid specifying the bold text directly.
bquote has a decent interface for this. You just surround the variable you want to substitute with .(). You could also use substitute with expression.
junk <- c("I'm Special", "You're Special")
plot(0, type="n")
mtext(bquote(paste("Do you think ", bold(.(junk[1])),"today?")),3,2)
mtext(bquote(paste("I think ", bold(.(junk[2])), "today.")),3,1)
Piston_Rings<-diameter[1:25,]
I want my quality control graph NOT to have the underscore in the object name.
At the moment there is an underscore (not a hyphen) in that object name. It is possible to construct objects whose names have spaces in them but in order to access them you will then always need to use backticks in order to get the interpreter to understand what you want:
> `Piston Rings` <- list(1,2)
> `Piston Rings`[[1]]
[1] 1
> `Piston Rings`[[2]]
[1] 2
The problem you incur is cluttering up your code, at least relative to obeying the usual conventions in R where a space is a token-ending marker to the parser. Hyphens (at least short-hyphens) are actually minus signs.
If on the other hand you only want to use a modified version of a name that contains an underscore as the title for a graph, then try something like this:
Piston_Rings <- list() # just for testing purposes so there will be an object.
plot( 1:10,10:1, main = sub("_", " ", quote(Piston_Rings)) )
#BondedDust's answer is correct, but (guessing, since you haven't been very specific) a simpler way to get what you want is just to specify xlab or ylab arguments to the plot() function. Let's say you have variables stuff (x) and Piston_Rings (y). If you just
plot(stuff,Piston_Rings)
then the plot will have "Piston_Rings" as the y-axis label. But if you
plot(stuff,Piston_Rings,ylab="Piston Rings")
you'll get the label you want. You can also include lots more information this way:
plot(stuff,Piston_Rings,
xlab="Important stuff (really)",
ylab="Piston Rings (number per segment)")
See ?plot.default for many more options.
I would like to have the title for the plot in two lines, but this does not work, why? and how can I make it work?
CVal<-1
SumEpsVal<-2
plot(1:10, main=bquote(paste("C=", .(CVal), " \n ", sum(xi), "=", .(SumEpsVal) )))
This here works:
plot(1:10, main=paste("C=1", "\n", "SumXi=2"))
I guess bquote makes something wrong... (look up ?bquote)
I tried to change environment in bqoute (the where-argument) but I don't know which environment to take.
BTW:
plot(1:10, main=bquote(paste("C=", .(CVal), "bla \n ", sum(xi), "=", .(SumEpsVal) )))
makes something crazy with the "bla".
Personally I would use mtext as already suggested. But if you really want it to be a one-liner, you can "cheat" bquote by using atop:
plot(1:10, main=
bquote(atop(paste("C=",.(CVal)), paste(sum(xi),"=",.(SumEpsVal)))))
It even aligns both lines neatly to the center.
The root issue is that plotmath does not support newlines within the
expressions to be output.
Control characters (e.g. \n) are not interpreted in character strings in plotmath,
unlike normal plotting.
You really need to create and output each line separately.
For example :
Lines <- list(bquote(paste("C=", .(CVal))),
bquote(paste(sum(xi), "=", .(SumEpsVal))))
Now output each line The text in the list is converted to expressions do.call
mtext(do.call(expression, Lines),side=3,line=0:1)
One way to achieve this is to use mtext to add an additional line under the main title as follows:
plot(1:10, main=bquote(paste("C=", .(CVal))))
mtext(bquote(paste(sum(xi), "=", .(SumEpsVal) )),side=3,line=0)
There may be a prettier solution, but perhaps this is enough for your needs.