如何在 R 中在月份数据之上绘制天数数据

Question

我在 R 中有日期数据，列出了 11 个月内的特定日期，看起来像这样（除了我有更多的数据点）：

Intake.Date = c("2023-01-26", "2023-01-26", "2023-01-26", "2023-01-26", "2023-01-26",
"2023-01-26", "2023-01-26","2023-02-02", "2023-01-31", "2023-02-07", "2023-01-31",
"2023-02-16", "2023-02-18", "2023-02-18","2023-03-02", "2023-03-02", "2023-03-02",
"2023-03-06", "2023-03-06", "2023-03-06", "2023-03-08")

我想绘制 2 层图表：一个条形图显示数据中每个月的计数，最上面显示每天的计数。我从这个开始：

ggplot()+
geom_histogram(data = AFOBC_Pathology_Data, mapping = aes(Intake.Date), bins = 11)+
geom_bar(data = AFOBC_Pathology_Data, mapping = aes(Intake.Date), fill = 'cyan')

产生以下图表：

这接近我想要的，但我知道垃圾箱（灰色条）不正确，因为我知道每个月的日期数量不同。但是，如果我尝试在同一张图表上绘制日期和月份，如下所示：

ggplot()+
geom_bar(data = AFOBC_Pathology_Data, mapping = aes(format(Intake.Date,'%b')), fill = 'grey')+
geom_bar(data = AFOBC_Pathology_Data, mapping = aes(Intake.Date), fill = 'cyan'),

我收到以下错误：

mapped_discrete()

错误：！无法将

x

转换为 . 运行

rlang::last_trace()

查看错误发生的位置。

Answer 1

第二种方法的问题是，在第一个

geom_bar

中，您映射了一个字符串（又名月份缩写，又名

x

上的离散变量），而在第二种方法中，它仍然是一个日期。

如果您想让条形显示该月的计数，同时也显示天数，我建议在

ggplot()

之外手动计算计数并使用

geom_rect

，例如类似的东西

library(ggplot2)
library(dplyr, warn=FALSE)
library(lubridate)

AFOBC_Pathology_Data_binned <- AFOBC_Pathology_Data |> 
  mutate(
    month = format(Intake.Date, "%b")
  ) |> 
  summarise(
    start = floor_date(min(Intake.Date), "month"), 
    end = ceiling_date(max(Intake.Date), "month"),
    count = n(),
    .by = "month"
  )


ggplot(AFOBC_Pathology_Data) +
  geom_rect(
    data = AFOBC_Pathology_Data_binned,
    aes(
      xmin = start, xmax = end,
      ymin = 0, ymax = count
    ),
    fill = "grey"
  ) +
  geom_bar(
    aes(Intake.Date),
    fill = "cyan"
  )

数据

AFOBC_Pathology_Data <- data.frame(
  Intake.Date = c(
    "2023-01-26", "2023-01-26", "2023-01-26", "2023-01-26", "2023-01-26",
    "2023-01-26", "2023-01-26", "2023-02-02", "2023-01-31", "2023-02-07", "2023-01-31",
    "2023-02-16", "2023-02-18", "2023-02-18", "2023-03-02", "2023-03-02", "2023-03-02",
    "2023-03-06", "2023-03-06", "2023-03-06", "2023-03-08"
  )
)
AFOBC_Pathology_Data$Intake.Date <- as.Date(AFOBC_Pathology_Data$Intake.Date)