conjure.simulation_machine: Simulate Claim Histories

In kasaai/simulationmachine: Individual Claims History Simulation Machine

Description

Simulate Claim Histories

Usage

## S3 method for class 'simulation_machine'
conjure(charm, seed = NULL,
  seed_features = seed, rows_per_partition = NULL, ...)

Arguments

charm	An charm object returned by simulation_machine().
seed	Seed for random number generation.
seed_features	(Optional) For backwards compatibility; see Details.
rows_per_partition	(Optional, currently unused) For parallel processing, number of observations that are treated at the same time.
...	Optional additional arguments, currently unused.

Details

In the original Simulation Machine paper, a second "seed" parameter is available to be set that controls the randomness of the feature generation stage of the simulation. The seed_features parameter is included for users who wish to reproduce the datasets generated by the original accompanying code. However, for most usage this parameter can be ignored, and is set equal to the seed argument by default.

Examples

charm <- simulation_machine(
  num_claims = 5000, 
  lob_distribution = c(0.25, 0.25, 0.30, 0.20), 
  inflation = c(0.01, 0.01, 0.01, 0.01), 
  sd_claim = 0.85, 
  sd_recovery = 0.85
)

conjure(charm, seed = 100)

kasaai/simulationmachine documentation built on Nov. 4, 2019, 3:31 p.m.