whep (version 0.3.0)

fill_linear: Fill gaps by linear interpolation, or carrying forward or backward.

Description

Fills gaps (NA values) in a time-dependent variable by linear interpolation between two points, or carrying forward or backwards the last or initial values, respectively. It also creates a new variable indicating the source of the filled values.

Usage

fill_linear(
  data,
  value_col,
  time_col = year,
  interpolate = TRUE,
  fill_forward = TRUE,
  fill_backward = TRUE,
  value_smooth_window = NULL,
  .by = NULL
)

Value

A tibble data frame (ungrouped) where gaps in value_col have been filled, and a new "source" variable has been created indicating if the value is original or, in case it has been estimated, the gapfilling method that has been used.

Arguments

data

A data frame containing one observation per row.

value_col

The column containing gaps to be filled.

time_col

The column containing time values. Default: year.

interpolate

Logical. If TRUE (default), performs linear interpolation.

fill_forward

Logical. If TRUE (default), carries last value forward.

fill_backward

Logical. If TRUE (default), carries first value backward.

value_smooth_window

An integer specifying the window size for a centered moving average applied to the variable before gap-filling. Useful for variables with high inter-annual variability. If NULL (default), no smoothing is applied.

.by

A character vector with the grouping variables (optional).

Examples

sample_tibble <- tibble::tibble(
  category = c("a", "a", "a", "a", "a", "a", "b", "b", "b", "b", "b", "b"),
  year = c(
    "2015", "2016", "2017", "2018", "2019", "2020",
    "2015", "2016", "2017", "2018", "2019", "2020"
  ),
  value = c(NA, 3, NA, NA, 0, NA, 1, NA, NA, NA, 5, NA),
)
fill_linear(sample_tibble, value, .by = c("category"))
fill_linear(
  sample_tibble,
  value,
  interpolate = FALSE,
  .by = c("category"),
)