如何用ggplot绘制自定义中位数与范围的多箱线图？

我希望在一张图表中绘制自定义中位数与数值范围的箱线图，但当前ggplot代码生成的图表不符合预期。目标是每个类别对应一个箱线图，且箱线的最小值、下四分位、中位数、上四分位、最大值都由数据框中指定的列控制。

现有问题代码 #

ggplot(data = df_subset[df, aes(y = target_fed_funds_rate_median, x = Date, group = Date)) +
  geom_boxplot(aes(ymin = target_fed_funds_rate_lr, lower = target_fed_funds_rate_cl, middle = target_fed_funds_rate_median,
    upper = target_fed_funds_rate_ch, ymax = target_fed_funds_rate_hr))

可复现代码 #

time = c(1, 2, 3, 4, 5)
median = c(1, 2, 3, 4, 5)
low = c(0.5, 1.5, 2.5, 3.5, 4.5)
high = c(1.5, 2.5, 3.5, 4.5, 5.5)
center_high = c(1.25, 2.25, 3.25, 4.25, 5.25)
center_low = c(0.75, 1.75, 2.75, 3.75, 4.75)

df = data.frame(median, low, high, center_high, center_low)

ggplot(data = df, aes(y = median, x = time, group = time)) +
  geom_boxplot(aes(ymin = low, lower = center_low, middle = median,
    upper = center_high, ymax = high))

修正方案 #

问题核心是geom_boxplot默认会自动计算统计量，覆盖手动指定的参数值，需添加stat = "identity"强制使用自定义统计量。

修正后的可复现代码 #

time = c(1, 2, 3, 4, 5)
median = c(1, 2, 3, 4, 5)
low = c(0.5, 1.5, 2.5, 3.5, 4.5)
high = c(1.5, 2.5, 3.5, 4.5, 5.5)
center_high = c(1.25, 2.25, 3.25, 4.25, 5.25)
center_low = c(0.75, 1.75, 2.75, 3.75, 4.75)

df = data.frame(median, low, high, center_high, center_low)

ggplot(data = df, aes(x = factor(time), group = time)) +
  geom_boxplot(
    aes(
      ymin = low, 
      lower = center_low, 
      middle = median, 
      upper = center_high, 
      ymax = high
    ),
    stat = "identity"  # 关键：使用手动指定的统计量，不自动计算
  ) +
  labs(x = "time", y = "value")

针对原始数据的修正代码 #

# 修正语法错误+添加stat参数
ggplot(data = df_subset, aes(x = Date, group = Date)) +
  geom_boxplot(
    aes(
      ymin = target_fed_funds_rate_lr, 
      lower = target_fed_funds_rate_cl, 
      middle = target_fed_funds_rate_median, 
      upper = target_fed_funds_rate_ch, 
      ymax = target_fed_funds_rate_hr
    ),
    stat = "identity"
  )

额外说明 #

• stat = "identity"是必须项：告诉ggplot直接使用提供的ymin/lower/middle/upper/ymax列值，而非自动计算
• 原始代码中df_subset[df是语法错误，需直接使用df_subset作为数据源
• 若x轴变量是数值型（如示例中的time），转为因子factor(time)可让每个类别对应独立箱线，布局更符合预期

内容的提问来源于stack exchange，提问作者TRV