主页 / coding / 问题 / 78871727
Accepted
Hydro
Asked: 2024-08-14 23:19:02 +0800 CST

ggplot 在 R 中用阴影表示缺失的月份？

  • 772

更新： 向原始数据框添加不确定性界限，该界限应为图的一部分

我之前问过这个问题，但我再次询问是想看看是否还有其他方法可以在不连接线条的情况下遮蔽缺失月份的数据。

DF <- data.frame(Date = seq(as.Date("2000-01-01"),to = as.Date("2005-12-31"),by = "day"), 
                 Obs = runif(2192, 5,10),
                 Sim = runif(2192, 8,12),
                 Up = runif(2192,13,15),
                 Low = runif(2192,1,4)) %>% 
      filter(between(month(Date),3,10))

ggplot(DF, aes(x = Date))+
  geom_ribbon(aes(ymin = Up, ymax = Low), fill = "grey30", alpha = 0.5)+
  geom_line(aes(y = Obs, col = "blue"))+
  geom_line(aes(y = Sim, col = "red"))

需要输出 这正是我想要的，但没有连接缺失月份的线路。

在此处输入图片描述

更新 2： 不确定性界限示例输出 在此处输入图片描述

1 个回答

  1. Best Answer

    我可能会使用这些数据来查找没有连续每日测量的任何时期的开始和停止时间：

    library(tidyverse)

missing_months <- DF %>% 
  mutate(missing = c(diff(Date), 1) > 1,
         missing = missing | lag(missing)) %>%
  filter(missing) %>%
  mutate(start_stop = rep(c("start", "stop"), length = n())) %>%
  select(Date, start_stop) %>%
  pivot_wider(names_from = start_stop, values_from = Date, values_fn = list) %>%
  unnest(cols = c(start, stop)) %>%
  mutate(start = start + 1, stop = stop - 1)

missing_months
#> # A tibble: 5 x 2
#>   start      stop      
#>   <date>     <date>    
#> 1 2000-11-01 2001-02-28
#> 2 2001-11-01 2002-02-28
#> 3 2002-11-01 2003-02-28
#> 4 2003-11-01 2004-02-29
#> 5 2004-11-01 2005-02-28

    为了防止线条连接在一起，您需要标记每个连续运行，并将此标签用作绘图中的分组变量。请记住，将数据透视为长格式将更容易绘图：

    DF %>%
  pivot_longer(c('Obs', 'Sim')) %>%
  group_by(name) %>%
  mutate(run = cumsum(c(0, diff(Date) > 1)), group = interaction(name, run)) %>%
  ggplot() +
  geom_rect(aes(xmin = start, xmax = stop, ymin = -Inf, ymax = Inf),
            data = missing_months, fill = "gray30", alpha = 0.5) +
  geom_ribbon(aes(group = group, x = Date, ymin = Low, ymax = Up), 
              fill = 'gray80') +
  geom_line(aes(x = Date, y = value, color = name, group = group), 
            linewidth = 0.1) +
  scale_color_manual(NULL, values = c("red3", "blue3", "green4", "black"),
                     guide = guide_legend(override.aes = list(linewidth = 1))) +
  theme_bw(20)

    在此处输入图片描述

    • 1

相关问题