X-Y Axis Formatting in R

Question

I am trying to plot a graph and format the y-axis such that the numbers have proper spacing (5,000 apart) and are formatted to have commas (eg 350,000). Also I am trying to get the x-axis (the date) to show more refinement such that I have more dates on the x-axis for the viewer to see. Every time I plot the graph I get numbers on the x-axis, not dates.

This is what I have so far

plot(date,sales,type="s",ylab="Sales",yaxt="n")
axis(1, at = seq(min(date), max(date), length.out = 1)
axis(2, at = seq(min(sales), max(sales), length.out = 20), labels = formatC(seq(min(revenue), max(revenue), length.out = 20),small.mark = " ", format = "d"), las = 2)

Here is the data I am using:

sales <- c(76103, 57300, 49875, 52113, 47891, 43531, 50909, 54182, 55884, 
63780, 57165, 59841, 65952, 67602, 70693, 76375, 83365, 82051, 
88568, 100717, 99344, 88980, 99034, 99593, 110497, 87223, 98350, 
102337, 116642, 116854, 138072, 137737, 84696, 64028, 74457, 
82260, 89841, 90251, 92486, 95298, 105186, 114004, 125486, 125330, 
121609, 124053, 127363, 115706, 115173, 108807, 106469, 112372, 
110860, 106773, 111647, 107490, 86029, 67618, 74113, 67344)

date <- structure(c(11292, 11382, 11474, 11566, 11657, 11747, 11839, 
11931, 12022, 12112, 12204, 12296, 12387, 12478, 12570, 12662, 
12753, 12843, 12935, 13027, 13118, 13208, 13300, 13392, 13483, 
13573, 13665, 13757, 13848, 13939, 14031, 14123, 14214, 14304, 
14396, 14488, 14579, 14669, 14761, 14853, 14944, 15034, 15126, 
15218, 15309, 15400, 15492, 15584, 15675, 15765, 15857, 15949, 
16040, 16130, 16222, 16314, 16405, 16495, 16587, 16679), class = "Date")

I was able to get the y axis numbers to round using: axis(2, at = seq(round(min(revenue)/100)*100, round(max(revenue)/100)*100, length.out = 5), labels = formatC(seq(round(min(revenue)/100)*100, round(max(revenue)/100)*100, length.out = 5),small.mark = " ", format = "d"), las = 2) — jessica, Jan 10 '16 at 03:12
I am using sales and revenue interchangeably. But I used the round function to get it integers that are multiple of 100 or 1000. Now I am trying to get the comma to show up on the numbers in the y-axis and for the date to appear as a string "2005-03-01" not as "123562" — jessica, Jan 10 '16 at 03:16
Why would you choose length=1 for the locations of the date ticks? — IRTFM, Jan 10 '16 at 03:32
Do you now understand why that was not delivering the desired sequence? — IRTFM, Jan 10 '16 at 04:41

score 2 · Accepted Answer · answered Jan 10 '16 at 05:58

Unless your plot is enormous (much bigger than your screen), the number of labels you're asking for will be unreadable, because they'll overlay each other. Following the spirit of what you're asking for (y-ticks every 5000, commas in y labels, more x labels that make sense):

# enlarge left margin to accommodate transposed labels
par(mar = c(5.1, 6.1, 2.1, 2.1))

# a sequence for tick marks
y_seq <- seq((min(sales) %/% 5000) * 5000, 
            (max(sales) %/% 5000 + 1) * 5000,
            by = 5000)

# a sequence for labels
y_labs <- prettyNum(y_seq, big.mark = ',')
y_labs[seq(1, length(y_labs), 2)] <- NA

plot(date, sales, type="s", ylab=NA, xaxt = 'n', yaxt = 'n')

# `axis.Date` makes sequencing and formatting easy; adapt as you like
axis.Date(1, date, at = seq.Date(min(date), max(date), by = 'year'), format = '%Y')
axis(2, at = y_seq, labels = y_labs, las = 2)

# add y label in a readable location
title(ylab = 'Sales', mgp = c(4.5, 1, 0))

which produces

If you want more labels, the approach is pretty customizable from here.

One last point: While you can always fight base R graphics until they do what you want, sometimes the fight isn't worth it when ggplot2 has much better defaults.

Just two lines:

library(ggplot2)
qplot(date, sales, geom = 'step')

makes

Alistaire, I love you! Thank you so much! – jessica Jan 10 '16 at 23:27 — jessica, Jan 10 '16 at 23:27

X-Y Axis Formatting in R

1 Answers1