Mathematical Annotation in R

Description

If the text argument to one of the text-drawing functions (text, mtext, axis, legend) in R is an expression, the argument is interpreted as a mathematical expression and the output will be formatted according to TeX-like rules. Expressions can also be used for titles, subtitles and x- and y-axis labels (but not for axis labels on persp plots).

In most cases other language objects (names and calls, including formulas) are coerced to expressions and so can also be used.

Details

A mathematical expression must obey the normal rules of syntax for any R expression, but it is interpreted according to very different rules than for normal R expressions.

It is possible to produce many different mathematical symbols, generate sub- or superscripts, produce fractions, etc.

The output from demo(plotmath) includes several tables which show the available features. In these tables, the columns of grey text show sample R expressions, and the columns of black text show the resulting output.

The available features are also described in the tables below:

The supported ‘scalable delimiters’ are | ( [ {, lceil, lfloor and their right-hand versions. "." is equivalent to "": the corresponding delimiter will be omitted. Delimiter || is supported but has the same effect as |.

The symbol font uses Adobe Symbol encoding so, for example, a lower case mu can be obtained either by the special symbol mu or by symbol("m"). This provides access to symbols that have no special symbol name, for example, the universal, or forall, symbol is symbol("\042"). To see what symbols are available in this way use TestChars(font=5) as given in the examples for points: some are only available on some devices.

Note to TeX users: TeX's \Upsilon is Upsilon1, TeX's \varepsilon is close to epsilon, and there is no equivalent of TeX's \epsilon. TeX's \varpi is close to omega1. vartheta, varphi and varsigma are allowed as synonyms for theta1, phi1 and sigma1.

sigma1 is also known as stigma, its Unicode name.

Control characters (e.g., \n) are not interpreted in character strings in plotmath, unlike normal plotting.

The fonts used are taken from the current font family, and so can be set by par(family=) in base graphics, and gpar(fontfamily=) in package grid.

Note that bold, italic and bolditalic do not apply to symbols, and hence not to the Greek symbols such as mu which are displayed in the symbol font. They also do not apply to numeric constants.

Other symbols

On many OSes and some graphics devices many other symbols are available as part of the standard text font, and all of the symbols in the Adobe Symbol encoding are in principle available via changing the font face or (see ‘Details’) plotmath: see the examples section of points for a function to display them. (‘In principle’ because some of the glyphs are missing from some implementations of the symbol font.) Unfortunately, postscript and pdf have support for little more than European (not Greek) and CJK characters and the Adobe Symbol encoding (and in a few fonts, also Cyrillic characters).

On Unix-alikes:

In a UTF-8 locale any Unicode character can be entered, perhaps as a \uxxxx or \Uxxxxxxxx escape sequence, but the issue is whether the graphics device is able to display the character. The widest range of characters is likely to be available in the X11 device using cairo: see its help page for how installing additional fonts can help. This can often be used to display Greek letters in bold or italic.

In non-UTF-8 locales there is normally no support for symbols not in the languages for which the current encoding was intended.

On Windows:

Any Unicode character can be entered into a text string via a \uxxxx escape, or used by number in a call to points. The windows family of devices can display such characters if they are available in the font in use. This can often be used to display Greek letters in bold or italic.

A good way to both find out which characters are available in a font and to determine the Unicode number is to use the ‘Character Map’ accessory (usually on the ‘Start’ menu under ‘Accessories->System Tools’). You can also copy-and-paste characters from the ‘Character Map’ window to the Rgui console (but not to Rterm).

References

Murrell, P. and Ihaka, R. (2000). An approach to providing mathematical annotation in plots. Journal of Computational and Graphical Statistics, 9, 582–599. doi: 10.2307/1390947.

The symbol codes can be found in octal in the Adobe reference manuals, e.g. for Postscript https://www.adobe.com/content/dam/acom/en/devnet/actionscript/articles/PLRM.pdf or PDF https://www.adobe.com/devnet/acrobat/pdfs/pdf_reference_1-7.pdf and in decimal, octal and hex at http://www.stat.auckland.ac.nz/~paul/R/CM/AdobeSym.html.

See Also

demo(plotmath), axis, mtext, text, title, substitute quote, bquote

Examples

require(graphics)

x <- seq(-4, 4, len = 101)
y <- cbind(sin(x), cos(x))
matplot(x, y, type = "l", xaxt = "n",
        main = expression(paste(plain(sin) * phi, "  and  ",
                                plain(cos) * phi)),
        ylab = expression("sin" * phi, "cos" * phi), # only 1st is taken
        xlab = expression(paste("Phase Angle ", phi)),
        col.main = "blue")
axis(1, at = c(-pi, -pi/2, 0, pi/2, pi),
     labels = expression(-pi, -pi/2, 0, pi/2, pi))


## How to combine "math" and numeric variables :
plot(1:10, type="n", xlab="", ylab="", main = "plot math & numbers")
theta <- 1.23 ; mtext(bquote(hat(theta) == .(theta)), line= .25)
for(i in 2:9)
    text(i, i+1, substitute(list(xi, eta) == group("(",list(x,y),")"),
                            list(x = i, y = i+1)))
## note that both of these use calls rather than expressions.
##
text(1, 10,  "Derivatives:", adj = 0)
text(1, 9.6, expression(
 "             first: {f * minute}(x) " == {f * minute}(x)), adj = 0)
text(1, 9.0, expression(
 "     second: {f * second}(x) "        == {f * second}(x)), adj = 0)


plot(1:10, 1:10)
text(4, 9, expression(hat(beta) == (X^t * X)^{-1} * X^t * y))
text(4, 8.4, "expression(hat(beta) == (X^t * X)^{-1} * X^t * y)",
     cex = .8)
text(4, 7, expression(bar(x) == sum(frac(x[i], n), i==1, n)))
text(4, 6.4, "expression(bar(x) == sum(frac(x[i], n), i==1, n))",
     cex = .8)
text(8, 5, expression(paste(frac(1, sigma*sqrt(2*pi)), " ",
                            plain(e)^{frac(-(x-mu)^2, 2*sigma^2)})),
     cex = 1.2)

## some other useful symbols
plot.new(); plot.window(c(0,4), c(15,1))
text(1, 1, "universal", adj = 0); text(2.5, 1,  "\\042")
text(3, 1, expression(symbol("\042")))
text(1, 2, "existential", adj = 0); text(2.5, 2,  "\\044")
text(3, 2, expression(symbol("\044")))
text(1, 3, "suchthat", adj = 0); text(2.5, 3,  "\\047")
text(3, 3, expression(symbol("\047")))
text(1, 4, "therefore", adj = 0); text(2.5, 4,  "\\134")
text(3, 4, expression(symbol("\134")))
text(1, 5, "perpendicular", adj = 0); text(2.5, 5,  "\\136")
text(3, 5, expression(symbol("\136")))
text(1, 6, "circlemultiply", adj = 0); text(2.5, 6,  "\\304")
text(3, 6, expression(symbol("\304")))
text(1, 7, "circleplus", adj = 0); text(2.5, 7,  "\\305")
text(3, 7, expression(symbol("\305")))
text(1, 8, "emptyset", adj = 0); text(2.5, 8,  "\\306")
text(3, 8, expression(symbol("\306")))
text(1, 9, "angle", adj = 0); text(2.5, 9,  "\\320")
text(3, 9, expression(symbol("\320")))
text(1, 10, "leftangle", adj = 0); text(2.5, 10,  "\\341")
text(3, 10, expression(symbol("\341")))
text(1, 11, "rightangle", adj = 0); text(2.5, 11,  "\\361")
text(3, 11, expression(symbol("\361")))