双 x 轴直方图（图的顶部和底部）显示两个不同城市的温度分布

Question

我正在尝试创建一个具有 2 个 x 轴的图。我想要在这幅图中有两个直方图，一个在城市 A 的顶部（上下颠倒），一个在城市 B 的底部。我希望 x 轴以度为单位显示温度。

我尝试了几种不同的入门方法，但都失败了。本人能力有限，主要是修改网上复制的代码。

city<-c(A, B, A, B, A, B)
temp<-c(20, 25, 30, 35, 30, 40)
data<-data.frame(city, temp)

g.top <- ggplot(data, aes(x = temp)) +
  theme_bw() +
  theme(plot.margin = unit(c(1,5,-30,6),units="points"),
        axis.title.y = element_text(vjust =0.25)) +
  geom_histogram()
g.bottom <- ggplot(data, aes(x = temp)) +
  theme_bw() +
  theme(plot.margin = unit(c(0,5,1,1),units="points")) +
  geom_histogram()
grid.arrange(g.top,g.bottom, heights = c(1/5, 4/5))


p <- ggplot(data, aes(x = temp, y=city)) + geom_histogram(aes(y=variable))

当我运行第一个代码块时，它给了我两个单独的堆叠直方图并说：

stat_bin() using bins = 30. Pick better value with binwidth.

但无论如何它都没有给我第二个 x 轴。

第二个代码给我一个错误，说我不能以 y 审美使用 statbin，这对我来说很有意义，但我现在没有想法。

我真的很沮丧，非常非常感谢任何建议

Answer 1

这不一定是到达那里的最简单方法，但可以控制整个过程。我将其作为数据重塑练习来处理。首先我得到每个城市的每个温度的计数，然后计算出这些计数之间有多少 space，然后绘制一个正常堆叠 geom_col.

# Get the counts
library(tidyverse)
my_data_transform <- data %>%
  count(temp, city, name = "count")

# Take that, and append after it rows that measure the "space between" each city's counts at each temperature.
my_data_transform2 <- bind_rows(
  my_data_transform,
  my_data_transform %>% 
    count(temp, wt = count, name = "count") %>%
    mutate(city = " ",
           count = 15 - count)  # Adjust to taste
  ) %>%
  # For plotting, I want the "space between" rows to be plotted in between the two cities.
  #  This is accomplished by making the city an ordered factor and putting the blanks 2nd.
  mutate(city = city %>% fct_relevel(" ", after = 1))

# Plot as geom_col, specifying clear bars for the middle ones.
ggplot(my_data_transform2, aes(temp, count, fill = city)) +
  geom_col() +
  scale_fill_manual(values = c("black", NA, "black")) +
  scale_y_continuous(labels = NULL, name = "") +
  theme_minimal()

Answer 2

分别绘制直方图，每个城市一个：

顶部 y-axis 翻转，x-axis 标签在顶部。
下图一如往常
设置 x-axis 对 both 图设置相同的限制，因此当我们使用 cowplot:plotgrid[= 垂直对齐 x-axis 时27=] x-axis 将对齐。

参见下面的示例：

library(ggplot2) library(cowplot) #bigger example data set.seed(1); data <- data.frame(city = sample(LETTERS[1:2], 100, replace = TRUE), temp = sample(1:50, 100, replace = TRUE)) g.top <- ggplot(data[ data$city == "A", ], aes(x = temp)) + geom_histogram() + scale_x_continuous(limits = c(10, 50), position = "top") + scale_y_reverse() + theme_minimal() g.bottom <- ggplot(data[ data$city == "B", ], aes(x = temp)) + geom_histogram() + scale_x_continuous(limits = c(10, 50)) + theme_minimal() plot_grid(g.top, g.bottom, labels = c("City A", "City B"), nrow = 2, align = 'hv')

双 x 轴直方图（图的顶部和底部）显示两个不同城市的温度分布

Dual x-axis histogram (top and bottom of plot) showing distributions of temperature for two different cities

r

histogram

ggplot2

gridextra