simulate_claims: simulate_claims

In jimbrig/lossdevtapp: Actuarial Loss Development

simulate_claims

Description

A function to simulate *transactional* actuarial claims/loss data for Property Casualty Insurance.

Usage

simulate_claims(
  n_claims = 1000,
  start_date = "2015-01-01",
  end_date = Sys.Date(),
  seed = 12345,
  loss_distribution = "lnorm",
  params = list(mean_log = 7.5, sd_log = 1.5),
  status_prob_open = 0.96,
  cache = FALSE,
  ...
)

Arguments

n_claims	Numeric - Number of claims to be simulated.
start_date, end_date	Character/Date - Start and End dates for simulation to create claims within (experience_period).
seed	Numeric - the seed is used to isolate randomness during statistical simulations.
loss_distribution	Character - must be one of the distributions mentioned in the details below. Defaults to lognormal.
params	Parameters associated with the specified 'loss_distribution' in a list (i.e. 'list(mean_log = 7.5, sd_log = 1.5)' for lognormal distribution).
status_prob_open	Numeric - must be within '0 < x < 1' and represents probability a claim is open when running binomial simulations for claims' status.
cache	Boolean/Logical - enable caching?
...	If needed

Details

Severity/Loss Distributions: - Normal: 'norm' - Parameters are 'mean' and 'sd'. - Lognormal: 'lnorm' - Parameters are 'meanlog' and 'sdlog'. - Gamma: 'gamma' - Shape, Rate, Scale - LogGamma: 'lgamma' - Shapelog, Ratelog - Pareto: 'pareto' - Shape and Scale - Weibull: 'weibull' - Shape and Scale - Generalized Beta: 'genbeta' - Shape1, Shape2, Shape3, Rate, Scale

Value

The return value, if any, from executing the function.

jimbrig/lossdevtapp documentation built on Aug. 31, 2024, 8:38 a.m.