问题
I wish to calculate the difference between the current row and previous row, by groups.
x = data.table(a=c(15, 25, 10, 12), b = c(1,1,2,2))
> x
a b
1: 15 1
2: 25 1
3: 10 2
4: 12 2
> x[, c:= a - c(NA, a[.I-1]), by=b]
Warning messages:
1: In a - c(NA, a[.I - 1]) :
longer object length is not a multiple of shorter object length
2: In `[.data.table`(x, , `:=`(c, a - c(NA, a[.I - 1])), by = b) :
RHS 1 is length 3 (greater than the size (2) of group 2). The last 1 element(s) will be discarded.
What I wish to obtain is:
> x
a b c
1: 15 1 NA
2: 25 1 10
3: 10 2 NA
4: 12 2 2
回答1:
#Using development version of data.table
x[,value:=a-shift(a,1,type="lag"),by=b]
a b value
1: 15 1 NA
2: 25 1 10
3: 10 2 NA
4: 12 2 2
来源：https://stackoverflow.com/questions/30405416/r-data-table-how-to-use-previous-row-value-within-group