Bayesian quantile regression model for claim count data

Mohd Fadzli Mohd Fuzi, Abdul Aziz Jemain, Noriszura Ismail

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Quantile regression model estimates the relationship between the quantile of a response distribution and the regression parameters, and has been developed for linear models with continuous responses. In this paper, we apply Bayesian quantile regression model for the Malaysian motor insurance claim count data to study the effects of change in the estimates of regression parameters (or the rating factors) on the magnitude of the response variable (or the claim count). We also compare the results of quantile regression models from the Bayesian and frequentist approaches and the results of mean regression models from the Poisson and negative binomial. Comparison from Poisson and Bayesian quantile regression models shows that the effects of vehicle year decrease as the quantile increases, suggesting that the rating factor has lower risk for higher claim counts. On the other hand, the effects of vehicle type increase as the quantile increases, indicating that the rating factor has higher risk for higher claim counts.

Original languageEnglish
Pages (from-to)124-137
Number of pages14
JournalInsurance: Mathematics and Economics
Volume66
DOIs
Publication statusPublished - 1 Jan 2016

Fingerprint

Quantile Regression
Count Data
Regression Model
Quantile
Count
Siméon Denis Poisson
Regression
Negative Binomial
Insurance
Estimate
Linear Model
Count data
Quantile regression
Regression model
Decrease
Rating
Factors

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Economics and Econometrics
  • Statistics and Probability

Cite this

Bayesian quantile regression model for claim count data. / Fuzi, Mohd Fadzli Mohd; Jemain, Abdul Aziz; Ismail, Noriszura.

In: Insurance: Mathematics and Economics, Vol. 66, 01.01.2016, p. 124-137.

Research output: Contribution to journalArticle

@article{a640fb25aaf84273bc96dfc6f3c9e778,
title = "Bayesian quantile regression model for claim count data",
abstract = "Quantile regression model estimates the relationship between the quantile of a response distribution and the regression parameters, and has been developed for linear models with continuous responses. In this paper, we apply Bayesian quantile regression model for the Malaysian motor insurance claim count data to study the effects of change in the estimates of regression parameters (or the rating factors) on the magnitude of the response variable (or the claim count). We also compare the results of quantile regression models from the Bayesian and frequentist approaches and the results of mean regression models from the Poisson and negative binomial. Comparison from Poisson and Bayesian quantile regression models shows that the effects of vehicle year decrease as the quantile increases, suggesting that the rating factor has lower risk for higher claim counts. On the other hand, the effects of vehicle type increase as the quantile increases, indicating that the rating factor has higher risk for higher claim counts.",
author = "Fuzi, {Mohd Fadzli Mohd} and Jemain, {Abdul Aziz} and Noriszura Ismail",
year = "2016",
month = "1",
day = "1",
doi = "10.1016/j.insmatheco.2015.11.004",
language = "English",
volume = "66",
pages = "124--137",
journal = "Insurance: Mathematics and Economics",
issn = "0167-6687",
publisher = "Elsevier",

}

TY - JOUR

T1 - Bayesian quantile regression model for claim count data

AU - Fuzi, Mohd Fadzli Mohd

AU - Jemain, Abdul Aziz

AU - Ismail, Noriszura

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Quantile regression model estimates the relationship between the quantile of a response distribution and the regression parameters, and has been developed for linear models with continuous responses. In this paper, we apply Bayesian quantile regression model for the Malaysian motor insurance claim count data to study the effects of change in the estimates of regression parameters (or the rating factors) on the magnitude of the response variable (or the claim count). We also compare the results of quantile regression models from the Bayesian and frequentist approaches and the results of mean regression models from the Poisson and negative binomial. Comparison from Poisson and Bayesian quantile regression models shows that the effects of vehicle year decrease as the quantile increases, suggesting that the rating factor has lower risk for higher claim counts. On the other hand, the effects of vehicle type increase as the quantile increases, indicating that the rating factor has higher risk for higher claim counts.

AB - Quantile regression model estimates the relationship between the quantile of a response distribution and the regression parameters, and has been developed for linear models with continuous responses. In this paper, we apply Bayesian quantile regression model for the Malaysian motor insurance claim count data to study the effects of change in the estimates of regression parameters (or the rating factors) on the magnitude of the response variable (or the claim count). We also compare the results of quantile regression models from the Bayesian and frequentist approaches and the results of mean regression models from the Poisson and negative binomial. Comparison from Poisson and Bayesian quantile regression models shows that the effects of vehicle year decrease as the quantile increases, suggesting that the rating factor has lower risk for higher claim counts. On the other hand, the effects of vehicle type increase as the quantile increases, indicating that the rating factor has higher risk for higher claim counts.

UR - http://www.scopus.com/inward/record.url?scp=84949482075&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949482075&partnerID=8YFLogxK

U2 - 10.1016/j.insmatheco.2015.11.004

DO - 10.1016/j.insmatheco.2015.11.004

M3 - Article

VL - 66

SP - 124

EP - 137

JO - Insurance: Mathematics and Economics

JF - Insurance: Mathematics and Economics

SN - 0167-6687

ER -