Fill in Missing Values with Previous or Next Value

fill() fills missing value NA in selected columns using the value in the next or previous entry. This method is particularly useful in scenarios where data is recorded only when it changes and not repeated across consecutive entries.

e.g.1. fill() defaults to replacing missing data from top to bottom; that is, it fills missing values using the value from the previous non-missing entry in the column.

library(tidyr)library(dplyr)library(tibble)
# create a datasetsales <- tribble(  ~quarter, ~year, ~sales,  "Q1",     1998,  5.52,  "Q2",     NA,    1.72,  "Q3",     NA,    6.69,  "Q4",     NA,    4.74,  "Q1",     2000,  9.77,  "Q2",     NA,    9.68,  "Q3",     NA,    9.27,  "Q4",     NA,    5.96)
# fill in missing values downwards (the default if not specified)sales %>% fill(year, .direction = "down")

Output:

# A tibble: 8 × 3 quarter year sales <chr> <dbl> <dbl> 1 Q1 1998 5.52 2 Q2 1998 1.72 3 Q3 1998 6.69 4 Q4 1998 4.74 5 Q1 2000 9.77 6 Q2 2000 9.68 7 Q3 2000 9.27 8 Q4 2000 5.96

e.g.2. Use .direction = "up" to fill in missing values upwards, i.e., replacing NA with the nearest non-missing value below in the column.

# create a datasetpets <- tribble(  ~rank, ~pet_type, ~breed,  4L,    NA,       "French Bulldogs",  5L,    NA,       "Bulldogs",  6L,    "Dog",    "Beagles",  1L,    NA,       "Persian",  2L,    NA,       "Maine Coon",  6L,    "Cat",    "American Short")
# fill in missing values upwardspets %>% fill(pet_type, .direction = "up")

Output:

# A tibble: 6 × 3 rank pet_type breed <int> <chr> <chr> 1 4 Dog French Bulldogs 2 5 Dog Bulldogs 3 6 Dog Beagles 4 1 Cat Persian 5 2 Cat Maine Coon 6 6 Cat American Short

e.g.3. Use .direction = "downup" to fill missing values in both directions: preferentially downwards; if not applicable, then upward. Alternatively, you an use "updown" to switch the preference order. In addition, the filling can be also operated on a per group basis specified by group_by().

# create a datasetfruits <- tribble(  ~store, ~product, ~stock,  "A",    "Apple",   10,  "A",    "Banana",  NA,  "A",    "Orange",  20,  "B",    "Apple",   NA,  "B",    "Banana",  15,  "B",    "Orange",  NA,  "C",    "Apple",   NA,  "C",    "Banana",  NA,  "C",    "Orange",  30)
fruits %>%  # separately for each store:  group_by(store) %>%  # fill in missing values upwards preferentially, otherwise downwards  fill(stock, .direction = "updown")

Output:
# A tibble: 9 × 3
# Groups:   store [3]
  store product stock
  <chr> <chr>   <dbl>
1 A     Apple      10
2 A     Banana     20
3 A     Orange     20
4 B     Apple      15
5 B     Banana     15
6 B     Orange     15
7 C     Apple      30
8 C     Banana     30
9 C     Orange     30

Amazing eBook to learn ggplot2 FAST & EASY