Replication Code and Data for "Tail-Sensitive Insurance Pricing: An Economic Extension of the Esscher Principle"

Description

This dataset contains the replication code and processed data for the paper "Tail-Sensitive Insurance Pricing: An Economic Extension of the Esscher Principle" by Wen Limin and Li Dongyan. The package includes: (1) R scripts for data preprocessing and empirical analysis; (2) a processed dataset (data_ex.xlsx) used in the study; (3) instructions for reproducing all tables and figures reported in the paper. The processed dataset is derived from the publicly available dataset: Guillen, Montserrat; Bolancé, Catalina; Frees, Edward W.; Valdez, Emiliano A. (2021), "Insurance data for homeowners and motor insurance customers monitored over five years", Mendeley Data, V1, doi: 10.17632/vfchtm5y7j.1. The original data are publicly available without access restrictions. The provided scripts (data_preparation.R) generate the analysis dataset from the raw data. All results in the paper can be fully replicated by following the instructions in the README file.

Steps to reproduce

1. Download the raw dataset from: https://data.mendeley.com/datasets/vfchtm5y7j/1 2. Place the file "data_ex.csv" in the "data" folder of the replication package: ./data/data_ex.csv 3. Open R (version 4.0 or higher) and set the working directory to the root folder of the replication package: Example: setwd("path/to/replication_package") 4. Run the data preprocessing script to generate the analysis dataset: source("scripts/data_preparation.R") Output: ./data/data_ex.xlsx 5. Run the scripts to reproduce the tables: Example: source("scripts/Table1.R") source("scripts/Table2.R") ... source("scripts/Table12.R") 6. Run the scripts to reproduce the figures: Example: source("scripts/Figure1.R") source("scripts/Figure2.R") All results reported in the paper can be reproduced by following the above steps.

Institutions

• Guangzhou College of Commerce

  Guangdong, Guangzhou
• Jiangxi Normal University

  Jiangxi, Nanchang

Categories

Licence