https://github.com/kasaai/simulationmachine

Individual claims history simulation machine
https://github.com/kasaai/simulationmachine

actuarial-science neural-networks reserving rstats simulation

Last synced: about 2 months ago
JSON representation

Individual claims history simulation machine

• Host: GitHub
• URL: https://github.com/kasaai/simulationmachine
• Owner: kasaai
• License: other
• Created: 2019-08-19T19:41:59.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2019-09-25T04:40:50.000Z (over 5 years ago)
• Last Synced: 2024-08-13T07:15:33.651Z (6 months ago)
• Topics: actuarial-science, neural-networks, reserving, rstats, simulation
• Language: R
• Homepage: https://blog.kasa.ai/posts/simulation-machine/
• Size: 385 KB
• Stars: 19
• Watchers: 2
• Forks: 6
• Open Issues: 1
• Metadata Files:
  • Readme: README.Rmd
  • License: LICENSE.md

Awesome Lists containing this project

• jimsghstars - kasaai/simulationmachine - Individual claims history simulation machine (R)

README

        
---

output: github_document

---

```{r, include = FALSE}

knitr::opts_chunk$set(

  collapse = TRUE,

  comment = "#>",

  fig.path = "man/figures/README-",

  out.width = "100%"

)

```

# simulationmachine

[![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://www.tidyverse.org/lifecycle/#experimental)

[![Travis build status](https://travis-ci.org/kasaai/simulationmachine.svg?branch=master)](https://travis-ci.org/kasaai/simulationmachine)

This package implements the claims history simulation algorithm detailed in [*An Individual Claims History Simulation Machine*](https://www.mdpi.com/2227-9091/6/2/29) by Andrea Gabrielli and Mario V. Wüthrich. The goal is to provide an easy-to-use interface for generating claims data that can be used for loss reserving research.

## Installation

You can install the development version from [GitHub](https://github.com/) with:

``` r

# install.packages("remotes")

remotes::install_github("kasaai/simulationmachine")

```

## Example

First, we can specify the parameters of a simulation using `simulation_machine()`:

```{r}

library(simulationmachine)

charm <- simulation_machine(

  num_claims = 50000, 

  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

)

charm

```

Once we have the charm object, we can use `conjure()` to perform the simulation.

```{r, message=FALSE}

library(dplyr)

records <- conjure(charm, seed = 100)

glimpse(records)

```

Let's see how many claims we drew:

```{r}

records %>% 

  distinct(claim_id) %>% 

  count()

```

If you prefer to have each row of the dataset to correspond to a claim, you can simply pivot the data with tidyr:

```{r}

records_wide <- records %>% 

  tidyr::pivot_wider(

    names_from = development_year, 

    values_from = c(paid_loss, claim_status_open),

    values_fill = list(paid_loss = 0)

  )

glimpse(records_wide)

````

-----

Please note that this project is released with a [Contributor Code of Conduct](https://github.com/kasaai/community/blob/master/CODE_OF_CONDUCT.md). By participating in this project you agree to abide by its terms.