4 ggplot2

单独介绍一些关于ggplot绘图相关的命令：因为在R中绘图占据很重要的一部分内容。

4.1 `boxplot`

对于箱线图，若对于变量x为因子，会自动的创建箱线图。

plot(ToothGrowth$supp, ToothGrowth$len)

进一步想要去探究不同变量之间的关系：这里使用的形式和lm()函数中类似。

boxplot(len~supp+dose, data = ToothGrowth)

4.1.1 使用`ggplot`

library(ggplot2)
ggplot(ToothGrowth, aes(x = supp, y = len)) +
  geom_boxplot()

ggplot(ToothGrowth, aes(x = interaction(supp, dose), y = len)) +
  geom_boxplot()

其中的interaction()表示的是x变量之间的交互关系共同构成x变量。

4.2 曲线curve

curve(x^3 + x^2 - 5*x, from = -5, to = 5)

一些比较复杂的做法是通过划分数据点进一步通过geom_line实现。

fx<-function(x)
{
  y<-0.5*log((1-x)/x)
  d<-data.frame(x=x,y=y)
  return(d)
}
x<-seq(0.005,0.5,0.005)
d<-fx(x)
ggplot(d,aes(x,y))+geom_line(color="red")+xlab("x")+ylab("y=0.5*log((1-x)/x)")+theme_bw()

在ggplot中也包含这样的一个函数ggfun()专用于绘制曲线函数图像。

myfun <- function(x) {
  0.5*log((1-x)/x)
}
ggplot(data.frame(x = c(0, 0.5)), aes(x = x)) +
  stat_function(fun = myfun, geom = "line",color="red")+theme_bw()

4.3 柱状图

柱状图所对应的数据是当有一列X是每个柱的位置，而Y表示的是对应柱子的高度。

BOD

  Time demand
1    1    8.3
2    2   10.3
3    3   19.0
4    4   16.0
5    5   15.6
6    7   19.8

ggplot(BOD, aes(x = Time, y = demand)) +
  geom_col()

ggplot(BOD, aes(x = factor(Time), y = demand)) +
  geom_col()

4.3.1 分组

4.4 Line

ggplot(BOD, aes(x = Time, y = demand)) +
  geom_line() +
  ylim(0, max(BOD$demand))

ggplot(BOD, aes(x = Time, y = demand)) +
  geom_line() +
  expand_limits(y = 0)

4.4.1 添加点

ggplot(BOD, aes(x = Time, y = demand)) +
  geom_line() +
  geom_point()

但一些时候，我们需要对数据点的集中性进行描述，若缺乏这部分的描述我们并不能观测到相关的数据形态。

library(gcookbook) # Load gcookbook for the worldpop data set
library(patchwork)
data(worldpop)
ggplot(worldpop, aes(x = Year, y = Population)) +
  geom_line()

ggplot(worldpop, aes(x = Year, y = Population)) +
  geom_line() +
  geom_point()

似乎数据都是在0以后较为密集，同时上升的趋势近似于指数形态，可以考虑取对数来观测。

# Same with a log y-axis
ggplot(worldpop, aes(x = Year, y = Population)) +
  geom_line() +
  geom_point() +
  scale_y_log10()

4.4.2 多条线muliple line

library(gcookbook) # Load gcookbook for the tg data set
data(tg)
# Map supp to colour
ggplot(tg, aes(x = dose, y = length, colour = supp)) +
  geom_line()

将supp映射到linetype上。

# Map supp to linetype
ggplot(tg, aes(x = dose, y = length, linetype = supp)) +
  geom_line()

ggplot(tg, aes(x = dose, y = length)) +
  geom_line()

通过观测数据：

tg

  supp dose length
1   OJ  0.5  13.23
2   OJ  1.0  22.70
3   OJ  2.0  26.06
4   VC  0.5   7.98
5   VC  1.0  16.77
6   VC  2.0  26.14

增加了linetype也就是根据supp的数据类生成了两个数据框，再这个两个数据框分别绘制line。

ggplot(tg, aes(x = dose, y = length, shape = supp)) +
  geom_line(position = position_dodge(0.2)) +           # Dodge lines by 0.2
  geom_point(position = position_dodge(0.2), size = 4)  # Dodge points by 0.2

4.5 Area绘图

# Convert the sunspot.year data set into a data frame for this example
sunspotyear <- data.frame(
    Year     = as.numeric(time(sunspot.year)),
    Sunspots = as.numeric(sunspot.year)
  )

ggplot(sunspotyear, aes(x = Year, y = Sunspots)) +
  geom_area()

绘制一个面积绘图和曲线结合的方式：先建立一个函数，将定义域设置为[left,right]内，(exclude)y[x<left|x>right]<-NA。

dnorm_limit <- function(x) {
    y <- dnorm(x)
    y[x < 0  |  x > 2] <- NA
    return(y)
}

# ggplot() with dummy data
p <- ggplot(data.frame(x = c(-3, 3)), aes(x = x))

p +
  stat_function(fun = dnorm_limit, geom = "area", fill = "blue", alpha = 0.2) +
  stat_function(fun = dnorm)

4.6 散点图

散点图通常用于反映两个连续变量之间的关系，我们进一步可以去使用拟合直线来表示这两个变量之间的关系。

library(gcookbook) # Load gcookbook for the heightweight data set
library(dplyr)


Attaching package: 'dplyr'

The following objects are masked from 'package:stats':

    filter, lag

The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union

data("heightweight")

ggplot(heightweight, aes(x = ageYear, y = heightIn)) +
  geom_point()

4.6.1 调整散点

ggplot(heightweight, aes(x = ageYear, y = heightIn)) +
  geom_point(shape = 10)

参数shape是对于散点图内的形状进行调整；而size是对散点大小进行调整。

ggplot(heightweight, aes(x = ageYear, y = heightIn)) +
  geom_point(size = 1.5)

hw <- heightweight %>%
  mutate(weightgroup = ifelse(weightLb < 100, "< 100", ">= 100"))

# Specify shapes with fill and color, and specify fill colors that includes an empty (NA) color
ggplot(hw, aes(x = ageYear, y = heightIn, shape = sex, fill = weightgroup)) +
  geom_point(size = 2.5) +
  scale_shape_manual(values = c(21, 24)) +
  scale_fill_manual(
    values = c(NA, "black"),
    guide = guide_legend(override.aes = list(shape = 21))
  )

4.6.1.1 散点内类别

ggplot(heightweight, aes(x =ageYear,y = heightIn))+
  geom_point(aes(color = sex))

ggplot(heightweight, aes(x = ageYear, y = heightIn)) +
  geom_point(aes(shape = sex, colour = sex)) +
  scale_shape_manual(values = c(1,2)) +
  scale_colour_brewer(palette = "Set1")

ggplot(heightweight, aes(x = ageYear, y = heightIn, colour = weightLb)) +
  geom_point()

ggplot(heightweight, aes(x = ageYear, y = heightIn, size = weightLb)) +
  geom_point()

显然从图像中看出年龄和身高是正相关的。

ggplot(heightweight, aes(x = ageYear, y = heightIn)) +
  geom_point()+
  stat_smooth(method = lm, se = FALSE, colour = "red")

`geom_smooth()` using formula = 'y ~ x'

ggplot(heightweight, aes(x = ageYear, y = heightIn)) +
  geom_point()+
  stat_smooth(method = lm, se = TRUE, colour = "red")

`geom_smooth()` using formula = 'y ~ x'

ggplot(heightweight, aes(x = ageYear, y = heightIn)) +
  geom_point(color="gray")+
  stat_smooth(method = loess, se = T, colour = "blue")

`geom_smooth()` using formula = 'y ~ x'

ggplot(heightweight, aes(x = ageYear, y = heightIn, color = sex)) +
  geom_point()+
  geom_smooth(method = lm, se = TRUE, fullrange = TRUE)

`geom_smooth()` using formula = 'y ~ x'

ggplot(heightweight, aes(x = ageYear, y = heightIn, color = sex)) +
  geom_point()+
  geom_smooth(method = loess, se = TRUE, fullrange = TRUE)

`geom_smooth()` using formula = 'y ~ x'

Warning: Removed 5 rows containing missing values (`geom_smooth()`).

还有诸如glm等拟合方法：

library(MASS)


Attaching package: 'MASS'

The following object is masked from 'package:dplyr':

    select

The following object is masked from 'package:patchwork':

    area

biopsy_mod <- biopsy %>%
  mutate(classn = recode(class, benign = 0, malignant = 1))

ggplot(biopsy_mod, aes(x = V1, y = classn)) +
  geom_point(
    position = position_jitter(width = 0.3, height = 0.06),
    alpha = 0.4,
    shape = 21,
    size = 1.5
  ) +
  stat_smooth(method = glm, method.args = list(family = binomial))

`geom_smooth()` using formula = 'y ~ x'

ggplot(heightweight, aes(x = ageYear, y = heightIn)) +
  geom_point()+
  geom_smooth(method = "lm")+
  annotate("text",x = 16.5,y = 53,label = "r^2==0.42",parse=T)

`geom_smooth()` using formula = 'y ~ x'

其中的参数parse可以使得表达式以数学公式表示。

ggplot(faithful, aes(x = eruptions, y = waiting)) +
  geom_point() +
  geom_rug()

geom_rug()可以在坐标轴上表明散点出现的频率。

4.7 统计图层

美学映射是图形语法中重要的一个概念，变量映射到元素上，通过几何形状GEOM画出图形。

学习使用STAT的原因可以归结如此：

“Even though the data is tidy, it may not represent the values you want to display”

simple_data <- tibble(group = factor(rep(c("A", "B"), each = 15)),
                      subject = 1:30,
                      score = c(rnorm(15, 40, 20), rnorm(15, 60, 10)))
simple_data

# A tibble: 30 × 3
   group subject score
   <fct>   <int> <dbl>
 1 A           1  31.3
 2 A           2  36.5
 3 A           3  20.6
 4 A           4  25.8
 5 A           5  64.9
 6 A           6  46.2
 7 A           7  22.5
 8 A           8  21.3
 9 A           9  35.5
10 A          10  49.3
# … with 20 more rows

假定我们想要画出一个柱状图，一个柱子代表每一组的group，柱子高度代表均值。显然直接使用GEOM是没有对应的绘图函数实现，首先需要对原始数据进行操作才可实现。

simple_data %>%
  group_by(group)%>%
  summarize(
    mean_score = mean(score),
    .groups = 'drop'
  )%>%
  ggplot(aes(x = group, y = mean_score))+
  geom_col()

其中传递给ggplot()的

simple_data %>%
  group_by(group) %>% 
  summarize(
    mean_score = mean(score),
    .groups = 'drop' 
  )

# A tibble: 2 × 2
  group mean_score
  <fct>      <dbl>
1 A           34.6
2 B           60.4

再将数据变形获得误差棒

simple_data %>% 
  group_by(group) %>% 
  summarize(
    mean_score = mean(score),
    se = sqrt(var(score)/length(score)),
    .groups = 'drop'
  ) %>% 
  mutate(
    lower = mean_score - se,
    upper = mean_score + se
  )

# A tibble: 2 × 5
  group mean_score    se lower upper
  <fct>      <dbl> <dbl> <dbl> <dbl>
1 A           34.6  4.11  30.4  38.7
2 B           60.4  2.09  58.3  62.5

传递到ggplot:

simple_data %>% 
  group_by(group) %>% 
  summarize(
    mean_score = mean(score),
    se = sqrt(var(score)/length(score)),
    .groups = 'drop'
  ) %>% 
  mutate(
    lower = mean_score - se,
    upper = mean_score + se
  )%>%
  ggplot(aes(x= group,y = mean_score,ymin = lower, ymax = upper))+
  geom_errorbar()

再进行组合先前的数据：

simple_data %>% 
  group_by(group) %>% 
  summarize(
    mean_score = mean(score),
    se = sqrt(var(score)/length(score)),
    .groups = 'drop'
  ) %>% 
  mutate(
    lower = mean_score - se,
    upper = mean_score + se
  )%>%
  ggplot() +
  geom_col(
    aes(x = group, y = mean_score),
  ) +
  geom_errorbar(
    aes(x = group, y = mean_score, ymin = lower, ymax = upper),
  )

再完成这样的一个图形之后，我们会发现其完成的步骤是较为繁琐的：

simple_data %>% 
  ggplot(aes(group, score)) +
  stat_summary(geom = "bar") +
  stat_summary(geom = "errorbar")

No summary function supplied, defaulting to `mean_se()`
No summary function supplied, defaulting to `mean_se()`

4.7.1 `stat_summary`

使用stat_summary是工作中最为常用的方法，为理解它举一个例子：

一个测试数据：

height_df <- tibble(group = "A",
                    height = rnorm(30, 170, 10))

height_df %>% 
  ggplot(aes(x = group, y = height)) +
  geom_point()

使用stat_summary代替geom_point

height_df %>% 
  ggplot(aes(x = group, y = height)) +
  stat_summary()

No summary function supplied, defaulting to `mean_se()`

最后会变成一条线和一个点，就像一个点的区间： geom_pointrange

height_df %>% 
  ggplot(aes(x = group, y = height))

height_df %>% 
  ggplot(aes(x = group, y = height)) +
  stat_summary(
    geom = "pointrange",
    fun.data = mean_se
  )

stat_summary函数可以进行调取geom_pointrange方法
参数fun.data 会调用函数将数据变形，这个函数默认是mean_se()
fun.data 返回的是数据框，这个数据框将用于geom参数画图，这里缺省的geom是pointrange
如果fun.data 返回的数据框包含了所需要的美学映射，图形就会显示出来。

4.1 boxplot

4.1.1 使用ggplot