臺灣汽車責任保險理賠之影響因素：貝氏迴歸模型

• Main Authors: Hsin-Yu Hsu, 許馨云
• Other Authors: Li-Hua Lai
• Format: Others
• Language: en_US
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/z46yxw

碩士 === 國立高雄第一科技大學 === 風險管理與保險研究所 === 104 === This study tries to combine rating factors and claim forecasting of automobile liability insurance, introducing human factors, marriage factor and vehicle factors and uses a property and casualty insurance company′s data, and adopts the Bayesian quantile regression and Negative binomial regression approaches to explore the effects of human factors, marriage factor and vehicle factors of claim frequency under high, middle and low claim quantile respectively. Thus, this study employ the method of Mean regression and Bayesian quantile regression approaches with middle claim frequency quantile to test the consistency of results from mean and middle values. The results of this study show that the low number of claims may represent as a good driving record of the policyholders. Therefore, the rating factors are not insignificantly on the low number of claims. Aged 26-30 and 31-60 are positive and significant on the middle and high number of claims. The vehicle c.c. is negative and significant on the middle and high number of claims. Except for human factors are used as the third party liability insurance rates by insurance companies, are generally, added on the marriage and vehicle factors will be used to determine the premium to be charged.