Neural network-based nonlinear prediction of reinsurance demand of the Chinese property insurance industry

2011 ◽  
Author(s):  
Tang tian ◽  
Wang ruopeng
Keyword(s):  
Neural Network ◽  
Insurance Industry ◽  
Nonlinear Prediction ◽  
Property Insurance

Identifying Key Fraud Indicators in the Automobile Insurance Industry Using SQL Server Analysis Services

2019 ◽  
Vol 64 (2) ◽  
pp. 53-71
Author(s):  
Botond Benedek ◽  
Ede László
Keyword(s):  
Neural Network ◽  
Decision Tree ◽  
Naive Bayes ◽  
Insurance Industry ◽  
Naïve Bayes ◽  
Sql Server ◽  
Categorical Variables ◽  
Automobile Insurance ◽  
Price Determination ◽  
Mining Tool

Abstract Customer segmentation represents a true challenge in the automobile insurance industry, as datasets are large, multidimensional, unbalanced and it also requires a unique price determination based on the risk profile of the customer. Furthermore, the price determination of an insurance policy or the validity of the compensation claim, in most cases must be an instant decision. Therefore, the purpose of this research is to identify an easily usable data mining tool that is capable to identify key automobile insurance fraud indicators, facilitating the segmentation. In addition, the methods used by the tool, should be based primarily on numerical and categorical variables, as there is no well-functioning text mining tool for Central Eastern European languages. Hence, we decided on the SQL Server Analysis Services (SSAS) tool and to compare the performance of the decision tree, neural network and Naïve Bayes methods. The results suggest that decision tree and neural network are more suitable than Naïve Bayes, however the best conclusion can be drawn if we use the decision tree and neural network together.

Configuration analysis of influencing factors of operating efficiency based on fsQCA: evidence from China’s property insurance industry

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Zhiguang Li ◽  
Yaokuang Li ◽  
Wei Zhang
Keyword(s):  
Technical Efficiency ◽  
Insurance Industry ◽  
Qualitative Comparative Analysis ◽  
Insurance Companies ◽  
Operating Efficiency ◽  
Content Type ◽  
Leverage Ratio ◽  
Property Insurance ◽  
High Loss ◽  
Core Conditions

Purpose Based on the perspective of complexity theory, the operation process of property insurance companies can be regarded as a complex dynamic nonlinear chaotic system. This paper aims to measure the operating efficiency of 29 Chinese domestic property and casualty (P&C) companies and 18 foreign-invested P&C companies from 2011 to 2017 and outline the path to achieving high-quality development. Design/methodology/approach The data were obtained from the Chinese Insurance Yearbook and China Statistical Yearbook 2012–2018. The data envelopment analysis method was used to calculate the technical efficiency of property insurance companies and fuzzy set qualitative comparative analysis is used for configuration analysis of determinants affecting technical efficiency. Findings This paper founds the average technical efficiency of Chinese domestic P&C insurance companies was 0.914 and that of foreign-invested P&C insurance companies was 0.895. The average total factor productivity of Chinese domestic P&C insurance companies was 1.058 and that of foreign-invested P&C insurance companies was 1.051. There were three modes to improve the company’s technical efficiency, with high loss ratio and low reinsurance ratio, poor employee education and higher leverage ratio and high leverage ratio and low reinsurance ratio as the core conditions. Originality/value This study puts forward four applicable, targeted and proven ways to improve the technical efficiency of China’s P&C insurance industry. These configurations were verified by the cases of existing property insurance companies, which can provide practical references for the insurance industry.

Knowledge Learning of Insurance Risks Using Dependence Models

2020 ◽  
Author(s):  
Zifeng Zhao ◽  
Peng Shi ◽  
Xiaoping Feng
Keyword(s):  
Large Scale ◽  
Predictive Analytics ◽  
Insurance Industry ◽  
Spatial Association ◽  
Capital Allocation ◽  
Risk Profile ◽  
Composite Likelihood ◽  
Multiple Sources ◽  
Property Insurance ◽  
Dependence Models

Learning the customers’ experience and behavior creates competitive advantages for any company over its rivals. The insurance industry is an essential sector in any developed economy and a better understanding of customers’ risk profile is critical to decision making in all aspects of insurance operations. In this paper, we explore the idea of using copula-based dependence models to learn the hidden risk of policyholders in property insurance. Specifically, we build a novel copula model to accommodate the dependence over time and over space among spatially clustered property risks. To tackle the computational challenge caused by the discreteness feature of large-scale insurance data, we propose an efficient multilevel composite likelihood approach for parameter estimation. Provided that latent risk induces correlation, the proposed customer learning method offers improved predictive analytics by allowing insurers to borrow strength from related risks in predicting new risks and also helps reveal the relative importance of the multiple sources of unobserved heterogeneity in updating policyholders’ risk profile. In the empirical study, we examine the loss cost of a portfolio of entities insured by a government property insurance program in Wisconsin. We find both significant temporal and spatial association among property risks. However, their effects on the predictive distribution of loss cost are different for the new and renewal policyholders. The two sources of dependence are complements for the former and substitutes for the latter. These findings are shown to have substantial managerial implications in key insurance operations such as experience rating, capital allocation, and reinsurance arrangement.

An integrated flood risk assessment model for property insurance industry in Taiwan

2011 ◽  
Vol 58 (3) ◽  
pp. 1295-1309 ◽  
Author(s):  
Wen-Ko Hsu ◽  
Pei-Chiung Huang ◽  
Ching-Cheng Chang ◽  
Cheng-Wu Chen ◽  
Dung-Moung Hung ◽  
...  
Keyword(s):  
Risk Assessment ◽  
Flood Risk ◽  
Insurance Industry ◽  
Assessment Model ◽  
Risk Assessment Model ◽  
Flood Risk Assessment ◽  
Property Insurance

scholarly journals An Empirical Study on the Equity Performance of China's Health Insurance Companies During the COVID-19 Pandemic—Based on Cases of Dominant Listed Companies

2021 ◽  
Vol 9 ◽  
Author(s):  
Rendao Ye ◽  
Na An ◽  
Yichen Xie ◽  
Kun Luo ◽  
Ya Lin
Keyword(s):  
Neural Network ◽  
Health Insurance ◽  
Stock Price ◽  
Short Term Memory ◽  
Insurance Industry ◽  
Back Propagation ◽  
Listed Companies ◽  
Insurance Companies ◽  
Neural Network Models ◽  
Health Insurance Companies

The health insurance industry in China is undergoing great shocks and profound impacts induced by the worldwide COVID-19 pandemic. Taking for instance the three dominant listed companies, namely, China Life Insurance, Ping An Insurance, and Pacific Insurance, this paper investigates the equity performances of China's health insurance companies during the pandemic. We firstly construct a stock price forecasting methodology using the autoregressive integrated moving average, back propagation neural network, and long short-term memory (LSTM) neural network models. We then empirically study the stock price performances of the three listed companies and find out that the LSTM model does better than the other two based on the criteria of mean absolute error and mean square error. Finally, the above-mentioned models are used to predict the stock price performances of the three companies.

Sign in / Sign up

Export Citation Format

Share Document