在R中绘制两个类的分布密度图

Question

我有一个包含两个类别（死亡和存活）的数据库，我想为每个类别的一个特定功能绘制一个分布图。

像这样： 能帮助吗？

可重现数据集：

dput(data[1:100,])

structure(list(`dead` = structure(c(1L, 1L, 1L, 
1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L), .Label = c("no", "yes"), class = "factor"), `Magnesium` = c(2.17, 
2.31, 2.14, 2.28, 2.17, 1.83, 1.9, 1.94, 2.68, 1.66, 2.14, 2.64, 
2.41, 2.19, 1.46, 2.34, 1.57, 2.52, 1.97, 1.8, 2.36, 1.6, 2.02, 
2.07, 1.44, 2.09, 2.08, 2.02, 2.52, 1.87, 2.72, 2.52, 1.58, 2.52, 
1.75, 2.81, 2, 1.83, 3.35, 1.74, 2.19, 2.44, 1.91, 2.33, 2.23, 
2.03, 2.13, 2.19, 2.02, 1.96, 2.52, 2.77, 2.17, 1.67, 2.04, 2.32, 
1.34, 1.75, 2.07, 2.23, 1.78, 2.69, 2.02, 3.1, 2.18, 1.61, 2.2, 
2.02, 2.3, 2, 2.45, 2.13, 1.96, 1.98, 2.1, 3.38, 1.36, 2.04, 
1.52, 3.12, 2.07, 2.68, 2.18, 2.59, 2.07, 1.77, 2.02, 2.31, 2.23, 
3.79, 1.41, 2.3, 1.97, 1.84, 1.95, 2.43, 2.17, 1.79, 1.7, 2.18
)), .Names = c("dead", "Magnesium"
), row.names = c(NA, 100L), class = "data.frame")

- anat

---

如果您能提供一个可重现的数据集和您已经尝试过的一些代码，那将会很有帮助。 - Adam Quek

@Adam Quek 我不知道该如何开始创建，因此我没有代码，但我已经添加了一个可重复的数据集。我遇到了一个不平衡的问题，所以是“yes”的数量不多。 - anat

Answer 1

d_list <- split(dat, dat$dead)
d_list <- lapply(d_list, function(x)density(x[, "Magnesium"])[c("x", "y")])
d_list <- lapply(d_list, function(x) do.call(data.frame, x))

plot(d_list[[1]], type="n", xlab="Magnesium", ylab="Density", main="Magnesium Level of two classes of patients")
for(i in seq(length(d_list))) lines(d_list[[i]], col=c("black", "red")[i], lty=i)
legend("topright", names(d_list), lty=1:2, col=c("black", "red"))

当然，在 ggplot 中这会更容易：

ggplot(dat, aes(Magnesium, col=dead, fill=dead)) + 
    geom_density(alpha=.5) 

- Adam Quek

Answer 2

使用 ggplot2（假设 df 是您的数据框）：

library(ggplot2)
ggplot(df) +
    geom_density(aes(x = Magnesium, 
                     group = dead, 
                     color = dead, 
                     fill = dead), 
                 alpha = .2, 
                 size = .25) +
    scale_fill_discrete('', labels = c('Alive', 'Dead')) +
    scale_color_discrete('', labels = c('Alive', 'Dead')) +
    scale_x_continuous(limits = c(0, NA)) +
    theme_minimal()

- GGamba