R 如何按年份划分变量的可用性？

那么2013年A组的积分应该是零，2014年A组的积分应该是零

---

您可以使用无范围的geom_errorbar（geom_errorbarh表示水平）。然后只需complete.cases（或

！is.na（df$value）

）的子集

库（ggplot2）
种子（10）
年份请不要使用我看到的df，谢谢你的提醒。y
ggplot（df）+几何图形条（aes（x=year，y=value，fill=group），position='dodge'，stat='identity'）
谢谢akrun，但这不是我想要的。价值不重要。我想展示的是每一组在哪一年有可用价值。对不起，我不确定我是否理解。谢谢。我认为我的描述令人困惑，我编辑了我的问题以作一些澄清。总体思路正确吗，但在缺少数据的地方，你想要的不是带分隔符的横线，而是横线？否则我还是不明白。我会说总的想法是正确的。在图形中，可用数据点显示为线段。但想象一下，将它们表示为点，并将这些点连接在同一组中。这些线可能需要连续多年的分组而不间断。这是可能的，但需要对数据做一些准备。但是，制作伪柱状图相当简单。请参阅我的编辑。
year <- c(2000:2014)
group <- c("A","A","A","A","A","A","A","A","A","A","A","A","A","A","A",
         "B","B","B","B","B","B","B","B","B","B","B","B","B","B","B",
         "C","C","C","C","C","C","C","C","C","C","C","C","C","C","C")
value <- sample(1:5, 45, replace=TRUE)

df <- data.frame(year,group,value)
df$value[df$value==1] <- NA

   year group value
1  2000     A    NA
2  2001     A     2
3  2002     A     2
...
11 2010     A     2
12 2011     A     3
13 2012     A     5
14 2013     A    NA
15 2014     A     3
16 2000     B     2
17 2001     B     3
...
26 2010     B    NA
27 2011     B     5
28 2012     B     4
29 2013     B     3
30 2014     B     5
31 2000     C     5
32 2001     C     4
33 2002     C     3
34 2003     C     4
...
44 2013     C     5
45 2014     C     3

2012 A 5
2013 A NA
2014 A 3

library(ggplot2)

set.seed(10)

year <- c(2000:2014)
group <- c("A","A","A","A","A","A","A","A","A","A","A","A","A","A","A",
       "B","B","B","B","B","B","B","B","B","B","B","B","B","B","B",
       "C","C","C","C","C","C","C","C","C","C","C","C","C","C","C")
value <- sample(1:5, 45, replace=TRUE)

df <- data.frame(year,group,value)
df$value[df$value==1] <- NA

no_na_df <- df[complete.cases(df), ]

ggplot(no_na_df, aes(x=year, y = group)) + 
    geom_errorbarh(aes(xmax = year, xmin = year), size = 2)

df$group_n <- as.numeric(df$group)

no_na_df <- df[complete.cases(df), ]

ggplot(no_na_df, aes(xmin=year-0.5, xmax=year+0.5, y = group_n)) + 
    geom_rect(aes(ymin = group_n-0.1, ymax = group_n+0.1)) +
    scale_y_discrete(limits = levels(df$group))