我正在尝试创建一个堆积条形图,但我无法定位图例,以便它不会与任何条形图重叠。我尝试使用单独的xlim
语句调整边距,设置legend
并在此处搜索,但都没有成功。实际上,示例数据集的代码来自StackOverflow。我使用基础R并且更喜欢基础R解决方案。谢谢你的任何建议。
setwd('c:/users/mmiller21/simple R programs/')
jpeg(filename = "barplot.jpeg")
set.seed(1234)
x <- replicate(8, round(10 * rexp(2000, 10)))
y <- apply(x, 2, function(column) table(factor(column, levels = 0:9)))
colnames(y) <- paste('A', seq(1,ncol(y),1), sep='')
rownames(y) <- paste('R', seq(1,nrow(y),1), sep='')
y
# A1 A2 A3 A4 A5 A6 A7 A8
# R1 769 800 790 806 792 787 834 801
# R2 779 733 793 757 786 744 731 776
# R3 284 297 278 272 263 301 280 275
# R4 112 106 91 124 106 103 104 96
# R5 33 38 37 26 36 37 30 36
# R6 11 18 7 11 10 20 11 9
# R7 8 8 3 2 3 3 9 5
# R8 4 0 1 2 4 4 0 2
# R9 0 0 0 0 0 1 1 0
# R10 0 0 0 0 0 0 0 0
par(mfrow=c(1, 1), mar=c(5, 5, 4, 2) + 0.1)
barplot(y, col = 1:nrow(y), ylab="My Variables", legend.text = TRUE,
args.legend = list(x = "topright", bty = "n"))
dev.off()
答案 0 :(得分:15)
这个怎么样:
library(RColorBrewer)
barplot(
y,
xlim=c(0, ncol(y) + 3),
col=brewer.pal(nrow(y), "Paired"),
ylab="My Variables",
legend.text=TRUE,
args.legend=list(
x=ncol(y) + 3,
y=max(colSums(y)),
bty = "n"
)
)
答案 1 :(得分:8)
有利润
par(mfrow=c(1, 1), mar=c(5, 5, 4, 8))
barplot(y, col = 1:nrow(y), ylab="My Variables", legend.text = TRUE,
args.legend = list(x = "topright", bty = "n", inset=c(-0.15, 0)))
答案 2 :(得分:6)
只是为了好玩,这里是ggplot2
,没有重叠的传说。
set.seed(1234)
x <- replicate(8, round(10 * rexp(2000, 10)))
y <- apply(x, 2, function(column) table(factor(column, levels = 0:9)))
y <- as.data.frame(y)
colnames(y) <- paste('A', seq(1,ncol(y),1), sep='')
rownames(y) <- paste('R', seq(1,nrow(y),1), sep='')
library(ggplot2)
library(reshape)
y$ID <- rownames(y)
y.melt <- melt(y, id.var = 'ID')
y.melt <- within(y.melt, ID <- factor(ID,
c('R10','R9','R8','R7','R6','R5','R4','R3','R2','R1'),
ordered = TRUE))
ggplot(y.melt, aes(x = variable, y = value, fill = ID)) +
geom_bar(stat = 'identity') +
xlab("") +
ylab("My variable") +
theme(legend.title=element_blank())
答案 3 :(得分:-1)
您应该在xpd=TRUE
par()
par(mfrow=.., mar=...,xpd=TRUE)