ggplot2根据数据拆分颜色或相邻直方图

Question

我如何创建直方图，其中 bin 在任意值之上和之下的颜色不同，而不需要引用 ggplot() 之外的环境？例如，像这样（但不需要引用外部变量

value.median

）：

value = stats::rnorm(100, mean = 0, sd = 1)

df = data.frame(value) %>%
  dplyr::mutate(above.median = value > median(value))

value.median = median(value)

df %>%
  ggplot(aes(x = value, fill = above.median)) +
  geom_histogram(boundary = value.median, alpha = 0.5)

不像这样（重叠的bin）：

df %>%
  ggplot(aes(x = value, fill = above.median)) +
  geom_histogram(boundary = 0, alpha = 0.5)

Answer 1

您可以将

position = "identity"

添加到geom，而不是默认的（即

position = "stack"

），例如

library(tidyverse)

set.seed(123)

df <- data.frame(value = rnorm(100, mean = 0, sd = 1))

df %>%
  mutate(above_below_zero = ifelse(value < median(value), "negative", "positive")) %>%
  ggplot(aes(x = value, fill = above_below_zero)) +
  geom_histogram(alpha = 0.5, bins = 30, position = "identity") +
  scale_fill_manual(values = c("negative" = "firebrick3",
                               "positive" = "deepskyblue"))

^{创建于 2023-03-13 与 reprex v2.0.2}