R Highcharter Errorbar 系列位置闪避

Question

我无法让误差线正确覆盖在我的分组列图上。在 ggplot 中我会使用 position = position_dodge()，但我无法在 highcharts 的错误栏文档中找到类似的内容。

这是一个 MWE：

df <- data.frame(Gender = c("Male", "Male", "Female", "Female"),
                 ShareType = rep(c("Long", "Short"),2),
                 InvestedPerAccount = c(10,9,7,8),
                 lower = c(8,7,6,7),
                 upper = c(11.5,10,9,8.8))



highchart() %>%
  hc_add_series(df, "column",hcaes(x = ShareType, y = InvestedPerAccount, group = Gender),
                tooltip = list(enabled = TRUE,pointFormat = '${point.y}')) %>%
  hc_add_series(df, 'errorbar', hcaes(x = ShareType, low = lower, high = upper, group = Gender, grouping = FALSE)) %>%
  hc_xAxis(categories = df$ShareType, title = list(text = "Share Type")) %>% 
  hc_colors(c("pink","lightblue"))

Answer 1

我不认为这是它应该工作的方式，但也许它是一种解决方法，直到你找到 函数。

highchart() %>%
  hc_add_series(df, "column",hcaes(x = ShareType, y = InvestedPerAccount, group = Gender),
                tooltip = list(enabled = TRUE,pointFormat = '${point.y}')) %>%
  hc_add_series(df, "errorbar", stemWidth = 1,  whiskerLength = 10, grouping = FALSE,
                centerInCategory = TRUE, groupPadding = .68,
                hcaes(x = ShareType, low = lower, high = upper, group = Gender)) %>%
  hc_xAxis(categories = df$ShareType, title = list(text = "Share Type")) %>% 
  hc_colors(c("pink","lightblue"))

Answer 2

@tamtam 的答案有效，但您需要将 groupPadding 设置为一个奇怪的恒定值，在本例中为 0.68。

真正的答案很简单 - 默认情况下，errorbar 系列连接到之前的系列。所以你的系列的顺序应该这样定义：

1. column 2. errorbar 3. column 4. errorbar

演示顺序正确：https://jsfiddle.net/BlackLabel/rmhw6f9y/

但 R 包装器按以下顺序添加它们：

1. column 2. column 3. errorbar 4. errorbar

演示顺序错误：https://jsfiddle.net/BlackLabel/bw0p68f3/

---

由于4.errorbar未连接到column（3.不是column series），errorbars分组错误。

解决方案很简单 - 您需要将 3. 与 1. 和 4. 与 2. 连接。为此，您可以使用 column.id 和 errorbar.linkedTo 属性：

https://api.highcharts.com/highcharts/series.column.id https://api.highcharts.com/highcharts/series.errorbar.linkedTo

我不知道如何将它们添加到图表的选项 (hc_add_series)，但我知道如何使用 JavaScript 更新系列：

hc_chart(events = list(load = JS("function () {
    this.series[0].update({
      id: 'firstColumnSeries'
    }, false);
    this.series[1].update({
      id: 'secondColumnSeries'
    }, false);
    this.series[2].update({
      linkedTo: 'secondColumnSeries'
    }, false);
    this.series[3].update({
      linkedTo: 'firstColumnSeries'
    });
  }"))) %>%

---

这是整个代码：

library('highcharter')

df <- data.frame(Gender = c("Male", "Male", "Female", "Female"),
                 ShareType = rep(c("Long", "Short"),2),
                 InvestedPerAccount = c(10,9,7,8),
                 lower = c(8,7,6,7),
                 upper = c(11.5,10,9,8.8))


highchart() %>%
  hc_chart(events = list(load = JS("function () {
    this.series[0].update({
      id: 'firstColumnSeries'
    }, false);
    this.series[1].update({
      id: 'secondColumnSeries'
    }, false);
    this.series[2].update({
      linkedTo: 'secondColumnSeries'
    }, false);
    this.series[3].update({
      linkedTo: 'firstColumnSeries'
    });
  }"))) %>%
  hc_add_series(df, "column",hcaes(x = ShareType, y = InvestedPerAccount, group = Gender),
                tooltip = list(enabled = TRUE,pointFormat = '${point.y}')) %>%
  hc_add_series(df, 'errorbar', hcaes(x = ShareType, low = lower, high = upper, group = Gender, grouping = FALSE)) %>%
  hc_xAxis(categories = df$ShareType, title = list(text = "Share Type")) %>% 
  hc_colors(c("pink","lightblue"))

ps。您可能希望将 secondColumnSeries 与 firstColumnSeries 交换，具体取决于您要放置错误栏的哪一侧。

Answer 3

解决方案是在第一组系列中使用id 参数，然后在错误栏系列中使用linkedTo。由于"errobar" 类型系列默认链接到上一个。

example_dat <- tibble(Type = c("Human", "High-Elf", "Orc"), 
       key = c("World1", "World2", "World3")) %>% 
  expand.grid() %>% 
  mutate(mean = runif(9, 5, 7)) %>% 
  mutate(sd = runif(9, 0, 2)) 


hchart(
  example_dat, 
  "column",
  hcaes(x = key, y = mean, group = Type),
  id = c("a", "b", "c")
  ) %>%
  
  hc_add_series(
    example_dat,
    "errorbar", 
    hcaes(y = mean, x = key, low = mean - sd, high = mean + sd, group = Type),
    linkedTo = c("a", "b", "c"),
    enableMouseTracking = TRUE,
    showInLegend = FALSE
    ) %>% 
  
  hc_plotOptions(
    errorbar = list(
      color = "black", 
      # whiskerLength = 1,
      stemWidth = 1
    ) 
  )