dplyr Window Functions Exercises in R: 30 Practice Problems
Thirty practice problems on dplyr window functions: lead, lag, cumulative ops, ranks, ntile, first/last/nth, cumall/cumany, with-groups patterns. Hidden solutions.
RRun this once before any exercise
library(dplyr)
library(tibble)
Section 1. lead and lag (8 problems)
Exercise 1.1: lag
Difficulty: Beginner.
Show solution
RInteractive R
tibble(x = 1:5) |> mutate(prev = lag(x))
Exercise 1.2: lead
Difficulty: Beginner.
Show solution
RInteractive R
tibble(x = 1:5) |> mutate(nxt = lead(x))
Exercise 1.3: lag with default
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(x = 1:5) |> mutate(prev = lag(x, default = 0))
Exercise 1.4: lag(n=2)
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(x = 1:5) |> mutate(prev2 = lag(x, n = 2))
Exercise 1.5: Day-over-day pct change
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(price = c(100, 102, 99, 105)) |>
mutate(pct = (price - lag(price)) / lag(price) * 100)
Exercise 1.6: lag per group
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(g = c("a","a","b","b"), v = c(1,2,3,4)) |>
group_by(g) |> mutate(prev = lag(v)) |> ungroup()
Exercise 1.7: Detect change-points
Difficulty: Advanced.
Show solution
RInteractive R
tibble(s = c("A","A","B","B","A")) |>
mutate(change = s != lag(s, default = first(s)))
Exercise 1.8: Days between events
Difficulty: Advanced.
Show solution
RInteractive R
tibble(d = as.Date(c("2024-01-01","2024-01-05","2024-01-15"))) |>
mutate(gap = as.integer(d - lag(d)))
Section 2. Cumulative (6 problems)
Exercise 2.1: cumsum
Difficulty: Beginner.
Show solution
RInteractive R
tibble(x = 1:5) |> mutate(cs = cumsum(x))
Exercise 2.2: cumprod
Difficulty: Beginner.
Show solution
RInteractive R
tibble(x = 1:5) |> mutate(cp = cumprod(x))
Exercise 2.3: cummax / cummin
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(x = c(3,1,5,4,8,2)) |> mutate(cmax = cummax(x), cmin = cummin(x))
Exercise 2.4: cummean
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(x = c(2,4,6,8)) |> mutate(cm = cummean(x))
Exercise 2.5: cumall
Difficulty: Advanced. Keep until first failure.
Show solution
RInteractive R
tibble(x = c(1,2,3,4,5,6)) |> filter(cumall(x < 5))
Exercise 2.6: cumany
Difficulty: Advanced. Keep AFTER first match.
Show solution
RInteractive R
tibble(x = c(1,2,3,4,5)) |> filter(cumany(x >= 3))
Section 3. Ranks (6 problems)
Exercise 3.1: row_number
Difficulty: Beginner.
Show solution
RInteractive R
tibble(x = c(50, 30, 70)) |> mutate(r = row_number(desc(x)))
Exercise 3.2: min_rank
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(x = c(50, 50, 70, 30)) |> mutate(r = min_rank(desc(x)))
Exercise 3.3: dense_rank
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(x = c(50, 50, 70, 30)) |> mutate(r = dense_rank(desc(x)))
Exercise 3.4: percent_rank
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(x = 1:10) |> mutate(pr = percent_rank(x))
Exercise 3.5: cume_dist
Difficulty: Advanced.
Show solution
RInteractive R
tibble(x = 1:5) |> mutate(cd = cume_dist(x))
Exercise 3.6: ntile
Difficulty: Intermediate. Quartiles of mpg.
Show solution
RInteractive R
mtcars |> mutate(quartile = ntile(mpg, 4)) |> count(quartile)
Section 4. first / last / nth (5 problems)
Exercise 4.1: first per group
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(g = c("a","a","b","b"), v = c(1,2,3,4)) |>
group_by(g) |> mutate(first_v = first(v)) |> ungroup()
Exercise 4.2: last per group
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(g = c("a","a","b","b"), v = c(1,2,3,4)) |>
group_by(g) |> mutate(last_v = last(v)) |> ungroup()
Exercise 4.3: nth value
Difficulty: Intermediate.
Show solution
RInteractive R
nth(c(10,20,30,40), n = 2)
Exercise 4.4: First non-NA
Difficulty: Advanced.
Show solution
RInteractive R
tibble(g = c("a","a","a"), v = c(NA, 5, 10)) |>
group_by(g) |> summarise(first_nonna = first(na.omit(v)))
Exercise 4.5: Days from first event
Difficulty: Advanced.
Show solution
RInteractive R
tibble(g = c("a","a","b","b"),
d = as.Date(c("2024-01-01","2024-02-15","2024-03-01","2024-03-10"))) |>
group_by(g) |>
mutate(days = as.integer(d - first(d))) |> ungroup()
Section 5. Group window patterns (5 problems)
Exercise 5.1: Top-N per group via row_number
Difficulty: Intermediate.
Show solution
RInteractive R
mtcars |> tibble::rownames_to_column("car") |>
group_by(cyl) |> mutate(rk = row_number(desc(mpg))) |>
filter(rk <= 2) |> ungroup()
Exercise 5.2: Cumulative spend per user
Difficulty: Intermediate.
Show solution
RInteractive R
tibble(u = c("a","a","b","b"), amt = c(50,80,30,70)) |>
group_by(u) |> mutate(cs = cumsum(amt)) |> ungroup()
Exercise 5.3: Running mean per group
Difficulty: Advanced.
Show solution
RInteractive R
tibble(u = c("a","a","a","b","b"), v = c(2,4,6,3,5)) |>
group_by(u) |> mutate(rm = cummean(v)) |> ungroup()
Exercise 5.4: Identify peaks
Difficulty: Advanced. Local max where x > lag(x) AND x > lead(x).
Show solution
RInteractive R
tibble(x = c(1, 5, 3, 8, 4, 9, 2)) |>
mutate(peak = x > lag(x, default = -Inf) & x > lead(x, default = -Inf))
Exercise 5.5: 3-period rolling mean via lags
Difficulty: Advanced.
Show solution
RInteractive R
tibble(x = c(10, 20, 30, 40, 50)) |>
mutate(ma3 = (x + lag(x) + lag(x, 2)) / 3)
What to do next
- dplyr-Exercises (shipped), broader practice.
- dplyr-Joins-Exercises (shipped), joins-only drills.