r - 如何过滤数据帧以获取连续增加的值

Question

我有一个包含多列的大型数据框，但对于此查询，我对 3 列感兴趣。

df <- structure(list(country = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = "ireland", class = "factor"), 
    parameter = structure(c(2L, 1L, 3L, 2L, 1L, 3L, 2L, 1L, 3L,2L, 1L, 3L), .Label = c("rainfall", "temp", "wind"), class = "factor"), 
    value = c(10L, 15L, 20L, 9L, 18L, 10L, 12L, 25L, 15L, 10L, 10L, 20L), unit = structure(c(3L, 2L, 1L, 3L, 2L, 1L, 3L, 
    2L, 1L, 3L, 2L, 1L), .Label = c("km/hr", "mm", "oC"), class = "factor")), class = "data.frame", row.names = c(NA, -12L))

country parameter   value   unit
ireland temp          10    oC
ireland rainfall      15    mm
ireland wind          20    km/hr
ireland temp           9    oC
ireland rainfall      18    mm
ireland wind          10    km/hr
ireland temp          12    oC
ireland rainfall      25    mm
ireland wind          15    km/hr
ireland temp          10    oC
ireland rainfall      10    mm
ireland wind          20    km/hr

我想按国家和参数分组以提取值列连续增加 3 次或更多次的行。
欲望输出的例子。

country parameter   value   unit
ireland rainfall      15    mm
ireland rainfall      18    mm
ireland rainfall      25    mm
ireland wind          10    km/hr
ireland wind          15    km/hr
ireland wind          20    km/hr

Answer 1

您可以按国家和参数分组，它们为不小于滞后值的值创建第三个分组变量，然后过滤组大小为 3 或更大:

library(dplyr)

df %>%
  group_by(country, parameter) %>%
  group_by(x = cumsum(value <= lag(value, default = FALSE)), .add = TRUE) %>%
  filter(n() >= 3) %>%
  ungroup() %>%
  arrange(country, parameter) %>%
  select(-x)

# A tibble: 6 x 4
  country parameter value unit 
  <fct>   <fct>     <int> <fct>
1 ireland rainfall     15 mm   
2 ireland rainfall     18 mm   
3 ireland rainfall     25 mm   
4 ireland wind         10 km/hr
5 ireland wind         15 km/hr
6 ireland wind         20 km/hr