
Generate a population with uniformly distributed ages
Source:R/generate_uniform_population.R
generate_uniform_population.RdCreate a starter population with independent uniformly distributed ages, randomly assigned sex, unknown parents, and no unions.
Usage
generate_uniform_population(
size,
max_age = 70,
end_time = 0,
min_age = 0,
female_probability = 0.5,
groups = NULL
)Arguments
- size
Number of living persons, a nonnegative whole number.
- max_age
Maximum initial age in years, finite and nonnegative.
- end_time
Reference time in years, finite. Birth times are this value minus the sampled ages. Defaults to a relative time origin of zero.
- min_age
Minimum initial age in years, finite and nonnegative, and no greater than
max_age. Defaults to zero.- female_probability
Probability of assigning sex "f" (female). The other sex value is "m" (male).
- groups
Optional group dimensions to sample, independently of age and sex. Either a named list with one named numeric vector of weights per dimension (categories as names; dimensions are drawn independently), or a data frame with one column per dimension and a numeric
weightcolumn, giving the joint distribution of category combinations. Weights are finite and nonnegative with at least one positive value, and are normalized.NULL(the default) adds no groups.
Details
Ages are sample from the uniform distribution between min_age and max_age
and only the living population is generated. This population will have no
father or mother ids. Sex and group categories follow their supplied
probabilities.
The generated population will not have a realistic age distribution. Typically, users will run a "burn in" simulation to produce a more realistic population distribution and kinship history for the desired simulation.
The end_time for the population will define the starting point for the
simulation. By default, this value is set to zero so that time is measured
since the beginning of the simulation. This means that birth times for the
existing population will be negative. Users who are drawing from input data
with time increments from real calendar time, should reset this starting
value as appropriate.
If the groups argument is used, the population declares each dimension with
group() defaults, except that its categories are those given, in order:
the weights' names
each dimension's values in order of first appearance in the table's rows.
The first is the reference category, and zero-weight categories are declared,
allowing for groups that may emerge later. set_groups() can replace that
declaration; categories it omits are kept.
Examples
set.seed(42)
# generate 50000 people with a maximum age of 50
population <- generate_uniform_population(size = 50000, max_age = 50)
# generate a population divided by race and region, with unequal division
# of race by region
region_race_dist <- tibble::tribble(
~race, ~region, ~weight,
"W", "NE", 0.145,
"W", "MW", 0.190,
"W", "S", 0.257,
"W", "W", 0.152,
"B", "NE", 0.022,
"B", "MW", 0.023,
"B", "S", 0.067,
"B", "W", 0.019,
"O", "NE", 0.021,
"O", "MW", 0.014,
"O", "S", 0.030,
"O", "W", 0.060,
)
population <- generate_uniform_population(50000, groups = region_race_dist)
prop.table(table(population$persons$race, population$persons$region))
#>
#> MW NE S W
#> B 0.02316 0.02238 0.06700 0.01818
#> O 0.01408 0.02154 0.03044 0.05972
#> W 0.18980 0.14254 0.25886 0.15230