complete.epi_df: "Complete" an 'epi_df', adding missing rows and/or replacing...

In cmu-delphi/epiprocess: Tools for basic signal processing in epidemiology

"Complete" an epi_df, adding missing rows and/or replacing NAs

Description

A tidyr::complete() analogue for epi_df objects. This function can be used, for example, to add rows for missing combinations of geo_value and time_value, filling other columns with NAs. See the examples for usage details.

Usage

## S3 method for class 'epi_df'
complete(data, ..., fill = list(), explicit = TRUE)

Arguments

data	an epi_df
...	see tidyr::complete
fill	see tidyr::complete
explicit	see tidyr::complete

Examples

start_date <- as.Date("2020-01-01")
daily_edf <- tibble::tribble(
  ~geo_value, ~time_value, ~value,
  1, start_date + 1, 1,
  1, start_date + 3, 3,
  2, start_date + 2, 2,
  2, start_date + 3, 3,
) %>%
  as_epi_df(as_of = start_date + 3)
# Complete without grouping puts all the geo_values on the same min and max
# time_value index
daily_edf %>%
  complete(geo_value, time_value = full_seq(time_value, period = 1))
# Complete with grouping puts all the geo_values on individual min and max
# time_value indices
daily_edf %>%
  group_by(geo_value) %>%
  complete(time_value = full_seq(time_value, period = 1))
# Complete has explicit=TRUE by default, but if it's FALSE, then complete
# only fills the implicit gaps, not those that are explicitly NA
daily_edf <- tibble::tribble(
  ~geo_value, ~time_value, ~value,
  1, start_date + 1, 1,
  1, start_date + 2, NA,
  1, start_date + 3, 3,
  2, start_date + 2, 2,
  2, start_date + 3, 3,
) %>%
  as_epi_df(as_of = start_date + 3)
daily_edf %>%
  complete(
    geo_value,
    time_value = full_seq(time_value, period = 1),
    fill = list(value = 0),
    explicit = FALSE
  )
# Complete works for weekly data and can take a fill value
# No grouping
weekly_edf <- tibble::tribble(
  ~geo_value, ~time_value, ~value,
  1, start_date + 1, 1,
  1, start_date + 15, 3,
  2, start_date + 8, 2,
  2, start_date + 15, 3,
) %>%
  as_epi_df(as_of = start_date + 3)
weekly_edf %>%
  complete(
    geo_value,
    time_value = full_seq(time_value, period = 7),
    fill = list(value = 0)
  )
# With grouping
weekly_edf %>%
  group_by(geo_value) %>%
  complete(
    time_value = full_seq(time_value, period = 7),
    fill = list(value = 0)
  )

cmu-delphi/epiprocess documentation built on Oct. 29, 2024, 5:37 p.m.