Skip to contents

Aggregate UCDP event data to common analytical units

Source: R/aggregate.R
ucdp_aggregate.Rd

ucdp_aggregate() constructs event counts and, when available, aggregated fatality bounds for country, PRIO-GRID, conflict, or dyad panels. Entity and time columns are detected from common UCDP names or can be supplied explicitly.

Usage

ucdp_aggregate(
  data,
  unit,
  id = NULL,
  date = NULL,
  by = NULL,
  include_fatalities = TRUE,
  low = NULL,
  best = NULL,
  high = NULL,
  na_rm = TRUE,
  missing = c("drop", "error")
)

Arguments

data

A data frame containing event-level or other disaggregated UCDP observations.

unit

One of "country-year", "country-month", "grid-year", "grid-month", "conflict-year", "conflict-month", "dyad-year", or "dyad-month". Underscores are accepted in place of hyphens.

id

Optional entity column or columns. When omitted, common names are detected for the selected unit.

date

Optional date column used to derive year and month. Existing year and month columns are preferred when both are available.

by

Additional grouping columns retained in the output.

include_fatalities

Include aggregated fatality estimates when a recognized triplet is present?

low, best, high

Optional fatality columns passed to the same detection logic as ucdp_fatalities().

na_rm

Missing-value behavior for fatality sums; see ucdp_fatalities().

missing

How to handle missing entity, time, or additional grouping values: drop their rows or raise an error.

Value

A tibble with one row per unit-period combination, n_events, and, when available, fatality totals and completeness counts. Provenance and an aggregation specification are attached as attributes.

Examples

events <- data.frame(
  country_id = c(1, 1, 1, 2), country = c("A", "A", "A", "B"),
  year = c(2020, 2020, 2021, 2020),
  low = c(0, 1, 2, 1), best = c(1, 2, 3, 2), high = c(2, 3, 4, 3)
)
ucdp_aggregate(events, unit = "country-year")
#> # A tibble: 3 × 9
#>   country_id country  year n_events n_fatality_complete n_fatality_missing
#>        <dbl> <chr>   <int>    <int>               <int>              <int>
#> 1          1 A        2020        2                   2                  0
#> 2          1 A        2021        1                   1                  0
#> 3          2 B        2020        1                   1                  0
#> # ℹ 3 more variables: fatalities_low <dbl>, fatalities_best <dbl>,
#> #   fatalities_high <dbl>