R：通过多个组对两个变量的滞后求和来填充 NA

Question

我尝试根据 N 的滞后和附加列的滞后来填充 N 列的 NA 值。问题是它填充了第二行，但不是全部。

这是我的代码。

```{r}
library(tidyverse)
library(zoo)

# toy data ----
df <- data.frame('Group.1' = c("a", "a", "a"),
                'Group.2' = c("A", "A", "A"),
                'monthyr' = as.yearmon(c("2018-01-01", "2018-02-01", "2018-03-01")),
                'x' = c(5, 7, 8),
                'y' = c(10, 18, 9),
                'N' = c(100, NA, NA)
                )
df$net <- df$x - df$y

df

# what i get ----
df_attempt <- df %>%
  group_by(Group.1, Group.2) %>%
  arrange(Group.1, Group.2, monthyr) %>%
  mutate(N = ifelse(is.na(N), lag(N) + lag(net), N))

# what i want to get ----
df_expected <- data.frame('Group.1' = c("a", "a", "a"),
                          'Group.2' = c("A", "A", "A"),
                          'monthyr' = as.yearmon(c("2018-01-01", "2018-02-01", "2018-03-01")),
                          'x' = c(5, 7, 8),
                          'y' = c(10, 18, 9),
                          'N' = c(100, 95, 84)
)
```

Answer 1

一种实现您尝试实现的“递归”填充的方法可能如下所示：

library(tidyverse)
library(zoo)

df %>%
  group_by(Group.1, Group.2) %>%
  arrange(monthyr) %>%
  fill(N) %>%
  mutate(N = N + cumsum(lag(net, default = 0)))
#> # A tibble: 3 × 7
#> # Groups:   Group.1, Group.2 [1]
#>   Group.1 Group.2 monthyr       x     y     N   net
#>   <chr>   <chr>   <yearmon> <dbl> <dbl> <dbl> <dbl>
#> 1 a       A       Jan 2018      5    10   100    -5
#> 2 a       A       Feb 2018      7    18    95   -11
#> 3 a       A       Mär 2018      8     9    84    -1

R：通过多个组对两个变量的滞后求和来填充 NA

R: Fill in NAs by multiple groups summing up lagges of two variables

r

dplyr

na

lag

mutated