An R Learner’s Diary: Base Package Graphics

Histograms

library(ggplot2)    # This package is used here only for its "diamonds" data set.
hist(x = diamonds$carat, main = "Carat Histogram", xlab = "Carat")

This shows the actual counts for each interval. For relative frequencies, use:

hist(x = diamonds$carat, main = "Carat Histogram", xlab = "Carat", freq = FALSE)

Scatter Plot

plot(x = diamonds$carat, y = diamonds$price)

The plot() function has the optional argument type. Possible types are:

• “p” for points,
• “l” for lines,
• “b” for both,
• “c” for the lines part alone of “b”,
• “o” for both ‘overplotted’,
• “h” for ‘histogram’ like (or ‘high-density’) vertical lines,
• “s” for stair steps,
• “S” for other steps, see ‘Details’ below,
• “n” for no plotting.

Other optional arguments include:

• main: an overall title for the plot.
• sub: a sub title for the plot.
• xlab: a title for the x axis.
• ylab: a title for the y axis.
• asp: the y/x aspect ratio.
• pch: the look of the dot. E.g. pch= “B”, pch=1, …, pch=18.
• lty: line type. E.g. lty=1, … , lty=6.
• lwd: line width. lwd=1 is standard.
• cex: a scaling factor for the sizes of the dots and text. cex=1 is standard.
• col: color. Can be specified in many ways.
• xlim = c(0, 100): specifies the x axis
• ylim = c(-100, 100): specifies the y axis
• las: label orientation. parallel to axes: las=0; horizontal: las=1; perpendicular to axes: las=2; vertical: las=3.

Note that the plotted variables can be from separate data frames.

Line graphs

x <- c(-0.5, 0, 1, 1, 1.5); y <- c( 0, 0, 2, 0, 0)
plot(x = x, y = y, lwd = 3, frame = FALSE, type = "l")

Next we add a polygon to the graph above

x <- c(-0.5, 0, 1, 1, 1.5); y <- c( 0, 0, 2, 0, 0)
plot(x = x, y = y, lwd = 3, frame = FALSE, type = "l")
polygon(x = c(0, .75, .75, 0), y = c(0, 0, 1.5, 0), lwd = 3, col = "lightblue")

Or we can fill the area under the graph:

x <- c(-0.5, 0, 1, 1, 1.5); y <- c( 0, 0, 2, 0, 0)
plot(x = x, y = y, lwd = 3, type = "l")
polygon(x = c(0, 1, 1, 0), y = c(0, 0, 2, 0), lwd = 3, col = "yellow")

Drawing Mathematical Functions as Curves: No Data Involved

This is done with the curve(function(x), xmin, xmax) command. Examples:

curve(x^2, -2, 2)

curve(dnorm(x), -3, 3)

Superimpose the Graphs, with add = TRUE

curve(x^2, -2, 2, lwd = 1, lty = 1)
curve(dnorm(x), add = TRUE, lwd = 1, lty = 2)

When one variable is plotted on the horizontal (x) axis, and multiple variables are plotted on the vertical (y) axis, we can use the matplot() command:

# Define one x vector for all:
year     <- c(2008,2009,2010,2011,2012,2013)
# Define a matrix of y values:
product1 <- c(0,3,6,9,7,8)
product2 <- c(1,2,3,5,9,6)
product3 <- c(2,4,4,2,3,2)
sales <- cbind(product1,product2,product3)
# plot
matplot(year,sales, type="b", lwd=c(1,2,3), col="black" )

Stack the graphs

par(mfrow=c(2,1))    # To properly stack the graphs
curve(x^2, -2, 2)
curve(dnorm(x), -3, 3)

The par(mfcol = c(rows,columns)) command could also be use to arrange multiple graphs.

Here’s a way to arrange three plots using the layout() function:

First, define a matrix:

(m <- matrix(c(1,1,2,3), 2, 2, byrow = TRUE))

##      [,1] [,2]
## [1,]    1    1
## [2,]    2    3

Then, use the matrix in the layout() function:

layout(matrix(c(1,1,2,3), 2, 2, byrow = TRUE))
# The first chart is sprawled across the first row
# The 2nd and 3rd plots are side by side in the second row
hist(mpg$displ)
hist(mpg$cty)
hist(mpg$hwy)

Adding Points, Lines, Arrows, and Text to a Plot

x <- c(1,3,4,7,8,9)
y <- c(0,3,6,9,7,8)
plot(x,y, main="Example for an Outlier")    # Just dots
points(8,1)                         # add a point
abline(a=0.31,b=0.97,lty=2,lwd=2)   # add a line; a=intercept; b=slope
text(7,2,"outlier",pos=3)           # add a label
arrows(7,2,8,1,length=0.15)         # add an arrow, from (7,2) to (8,1)

A Legendary Section

curve( dnorm(x,0,1), -10, 10, lwd=1, lty=1)
curve( dnorm(x,0,2),add=TRUE, lwd=2, lty=2)   
curve( dnorm(x,0,3),add=TRUE, lwd=3, lty=3)
legend("topright",c("sigma=1","sigma=2","sigma=3"), lwd=1:3, lty=1:3)

More sophisticated formatting:

curve( dnorm(x,0,1), -10, 10, lwd=1, lty=1)
curve( dnorm(x,0,2),add=TRUE, lwd=2, lty=2)
curve( dnorm(x,0,3),add=TRUE, lwd=3, lty=3)
legend("topleft",expression(sigma==1,sigma==2,sigma==3),lwd=1:3,lty=1:3)
## The expression next looks like LaTeX code without the backslashes
text(6,0.3, expression(f(x)==frac(1,sqrt(2*pi)*sigma)*e^{-frac(x^2,2*sigma^2)}))

Pie Charts and Bar Charts

Downloading and preparing the data:

# load data set
library(foreign)
affairs<-read.dta("http://fmwww.bc.edu/ec-p/data/wooldridge/affairs.dta")

# Generate "Factors" to attach labels
unique(affairs$kids)  # 0 1; shows the unique values in the variable

## [1] 0 1

haskids <- factor(affairs$kids,labels=c("no","yes"))
unique(affairs$ratemarr)  # 4 5 3 2 1; these 5 unique values can be converted to a factor

## [1] 4 5 3 2 1

mlab <- c("very unhappy","somewhat unhappy","average", "somewhat happy", "very happy")
marriage <- factor(affairs$ratemarr, labels=mlab)

Pie Chart

pie(table(marriage), col = gray(seq(0.2,1,0.2)))

Bar Plot (horizontal bars)

barplot(table(marriage), horiz = TRUE, las = 1, main = "Distribution of Happiness")

Stacked Bar Plot

barplot(table(haskids, marriage), horiz = TRUE, las = 1, legend = TRUE, args.legend = c(x="bottomright"), main = "Happiness by Kids")

Bunched Bars

barplot(table(haskids, marriage), beside = TRUE, las = 2, legend = TRUE, args.legend = c(x="top"), main = "Happiness by Kids")

This is it for now.