Question

我想比较不同时间条件之间的绝对差异。例如，这里有两个条件（A和B），每个条件在整个时间（1：3）中都有自己的值。

df <- data.frame("condition" =c("A","B","A","A","B","B"), "time" = c(1,1,2,3,2,3) ,"value" = c(-2,3,5,1,1,6))

我如何使用dplyr创建新条件（C）作为条件A和B之间的绝对差？

expected_outcome <- data.frame("condition" =c("A","B","A","A","B","B","Diff","Diff","Diff"), "time" = c(1,1,2,3,2,3,1,2,3) ,"value" = c(-2,3,5,1,1,6,1,4,5))

要使用这三个条件创建图？例如：

ggplot(a, aes(x=time,y=value,color= condition)) +stat_summary(fun.y="mean", geom="line")

任何建议都非常感谢！

Answer 1

模数列顺序：

df %>% spread(condition, value) %>%
    mutate(Diff=abs(abs(A) - abs(B))) %>%
    gather(condition, value, -time)

Answer 2

aggregate将达到目的：

dif <- aggregate(df$value, list(df$time), function(x) abs(diff(abs(x))))
dif <- data.frame(condition = "Diff", time = dif[,1], value = dif[,2])
dif <- rbind(df, dif)
dif

   condition time value
1         A    1    -2
2         B    1     3
3         A    2     5
4         A    3     1
5         B    2     1
6         B    3     6
7      Diff    1     1
8      Diff    2     4
9      Diff    3     5

使用dplyr计算不同时间条件值之间的绝对差

2 个答案: