slider::slide_dbl就是你要找的

slider::slide_dbl(myVec$myVec, sum,.before = 2, .after = 0)
[1]  7  9 13 11 10  9  6 14

字符串

下面是一个有效的向量化方法：

> v=c(7,2,4,5,1,3,2,9)
> o=outer(1:length(v),-2:0,"+")
> rowSums(matrix(v[ifelse(o<1|o>length(v),NA,o)],length(v)),na.rm=T)
[1]  7  9 13 11 10  9  6 14

字符串
基准：

v=sample(1:1e5)

msum=\(x,y){
  o=outer(1:length(x),y,"+")
  rowSums(matrix(x[ifelse(o<1|o>length(x),NA,o)],length(x)),na.rm=T)
}

nonvectorized=\(x,y){
  l=length(x)
  sapply(1:l,\(i)sum(x[max(1,i-y):i],na.rm=T))
}

microbenchmark::microbenchmark(times=10,
  vectorized=msum(v,-2:0),
  nonvectorized=nonvectorized(v,2),
  rollapplyr=zoo::rollapplyr(v,3,sum,partial=T),
  slide_dbl=slider::slide_dbl(v,sum,.before=2,.after=0)
)

Unit: milliseconds
          expr       min        lq      mean    median        uq       max neval
    vectorized  12.66232  17.63446  17.74140  17.86735  18.41807  25.08393    10
 nonvectorized 154.50835 163.04709 180.35591 172.72742 183.88661 252.43063    10
    rollapplyr 737.47610 789.48767 819.80156 797.42164 820.30966 973.17992    10
     slide_dbl  53.21695  55.42821  61.68474  58.34419  62.53458  83.75196    10

您可以在rollapplyr中使用partial=TRUE：

library(zoo)

rollapplyr(myVec$myVec, 3, sum, partial = TRUE)

# [1]  7  9 13 11 10  9  6 14

字符串
或者在dplyr中使用相同的函数：

library(dplyr)

myVec %>%
  mutate(myVec = rollapplyr(myVec, 3, sum, partial = TRUE))

型