scholarly journals Estimating Major Risk Factor Relativities in Rate Filings Using Generalized Linear Models

2018 ◽  
Vol 6 (4) ◽  
pp. 84 ◽  
Author(s):  
Shengkun Xie ◽  
Anna Lawniczak

Predictive modeling is a key technique in auto insurance rate-making and the decision-making involved in the review of rate filings. Unlike an approach based on hypothesis testing, the results from predictive modeling not only serve as statistical evidence for decision-making, they also discover relationships between a response variable and predictors. In this work, we study the use of predictive modeling in auto insurance rate filings. This is a typical area of actuarial practice involving decision-making using industry loss data. The aim of this study was to offer some general guidelines for using predictive modeling in regulating insurance rates. Our study demonstrates that predictive modeling techniques based on generalized linear models (GLMs) are suitable in auto insurance rate filings review. The GLM relativities of major risk factors can serve as the benchmark of the same risk factors considered in auto insurance pricing.

2021 ◽  
Author(s):  
Shengkun Xie ◽  
Anna T. Lawniczak

Predictive modeling is a key technique in auto insurance rate-making and the decision-making involved in the review of rate filings. Unlike an approach based on hypothesis testing, the results from predictive modeling not only serve as statistical evidence for decision-making, they also discover relationships between a response variable and predictors. In this work, we study the use of predictive modeling in auto insurance rate filings. This is a typical area of actuarial practice involving decision-making using industry loss data. The aim of this study was to offer some general guidelines for using predictive modeling in regulating insurance rates. Our study demonstrates that predictive modeling techniques based on generalized linear models (GLMs) are suitable in auto insurance rate filings review. The GLM relativities of major risk factors can serve as the benchmark of the same risk factors considered in auto insurance pricing.


2021 ◽  
Author(s):  
Shengkun Xie ◽  
Anna T. Lawniczak

Predictive modeling is a key technique in auto insurance rate-making and the decision-making involved in the review of rate filings. Unlike an approach based on hypothesis testing, the results from predictive modeling not only serve as statistical evidence for decision-making, they also discover relationships between a response variable and predictors. In this work, we study the use of predictive modeling in auto insurance rate filings. This is a typical area of actuarial practice involving decision-making using industry loss data. The aim of this study was to offer some general guidelines for using predictive modeling in regulating insurance rates. Our study demonstrates that predictive modeling techniques based on generalized linear models (GLMs) are suitable in auto insurance rate filings review. The GLM relativities of major risk factors can serve as the benchmark of the same risk factors considered in auto insurance pricing.


Risks ◽  
2021 ◽  
Vol 9 (7) ◽  
pp. 126
Author(s):  
Shengkun Xie

In insurance rate-making, the use of statistical machine learning techniques such as artificial neural networks (ANN) is an emerging approach, and many insurance companies have been using them for pricing. However, due to the complexity of model specification and its implementation, model explainability may be essential to meet insurance pricing transparency for rate regulation purposes. This requirement may imply the need for estimating or evaluating the variable importance when complicated models are used. Furthermore, from both rate-making and rate-regulation perspectives, it is critical to investigate the impact of major risk factors on the response variables, such as claim frequency or claim severity. In this work, we consider the modelling problems of how claim counts, claim amounts and average loss per claim are related to major risk factors. ANN models are applied to meet this goal, and variable importance is measured to improve the model’s explainability due to the models’ complex nature. The results obtained from different variable importance measurements are compared, and dominant risk factors are identified. The contribution of this work is in making advanced mathematical models possible for applications in auto insurance rate regulation. This study focuses on analyzing major risks only, but the proposed method can be applied to more general insurance pricing problems when additional risk factors are being considered. In addition, the proposed methodology is useful for other business applications where statistical machine learning techniques are used.


2014 ◽  
Vol 17 (4) ◽  
pp. 951-962 ◽  
Author(s):  
Eva Boj ◽  
Teresa Costa ◽  
Josep Fortiana ◽  
Anna Esteve

2020 ◽  
pp. 29-32
Author(s):  
V. A. Mishchenko ◽  
I. A. Kshnyasev ◽  
I. V. Vyalykh ◽  
I. P. Bykov ◽  
L. G. Vyatkina

The regions of the Ural Federal District (UFD) are highly endemic for tick-borne encephalitis (ТВE) territories. The dynamics of morbidity of ТВЕ in the population characterized by complex cyclic, depending on many external variables. Retrospective analysis of the long-term dynamics (2007–2019) of the incidence of ТВЕ in regions of the UFD, taking into account the number of tick affected people was presented. The chances of getting sick in tick affected peoples to quantify the effect of predictors on TBE incidence were calculated. Standard apparatus of the theory of generalized linear models – logit-regression was used. It was established that the regions of the UFD characterized by a similar dynamics in the odds ratio indicator, therefore, TBE incidence with alternating ups and downs with a trend towards a decrease in the chance of TBE getting sick in tick affected people from 2007 to 2019. On average, over 13 years, the chances of developing TBE are statistically significantly different in the studied regions of the UFD, which can be explained by the influence of many risk factors and their combinations on the TBE incidence.


Author(s):  
Łukasz Delong ◽  
Mathias Lindholm ◽  
Mario V. Wüthrich

AbstractThe most commonly used regression model in general insurance pricing is the compound Poisson model with gamma claim sizes. There are two different parametrizations for this model: the Poisson-gamma parametrization and Tweedie’s compound Poisson parametrization. Insurance industry typically prefers the Poisson-gamma parametrization. We review both parametrizations, provide new results that help to lower computational costs for Tweedie’s compound Poisson parameter estimation within generalized linear models, and we provide evidence supporting the industry preference for the Poisson-gamma parametrization.


Author(s):  
Lorilee A. Medders ◽  
Jamie Anderson-Parson ◽  
Matthew Thomas-Reid

There are three goals of insurance rate regulation. Rates must be: 1) adequate; 2) not excessive; and 3) not unfairly discriminatory. Rates that are adequate yet not excessive are overall high enough to pay claims and expenses, yet not so high overall that they result in unreasonable profiteering by insurers. The third regulatory goal—that rates are not unfairly discriminatory—is the topic of interest in our research. The concept of unfair discrimination in an insurance context—determining what constitutes fairness in pricing—can differ substantially from the thinking on fairness in a societal context. As a result, the term “discrimination” may be used quite differently in these two contexts. Discrimination, with negative societal connotations, is endemic in our world broadly and largely unjustifiable, yet in the narrower world of insurance, it is the basis for the entire industry’s viability and sustainability. In the insurance context, we can receive the term “discrimination” in a neutral manner, simply taking it to mean different treatment for different groups having different characteristics, without it necessarily connoting any negative intent or outcome. Indeed, the purpose in insurance for engaging in “fair discrimination” —that is, discrimination that price differentiates between discernibly different levels of risk—is itself rooted in economic fairness. An insurance carrier charges differential prices for its products based on differentials in risk. Nevertheless, when risk transfer to an insurer is priced based on uncontrollable and/or immutable classifications such as race and gender, there can be profoundly different views of what constitutes fairness. In many areas of U.S. law, discrimination on either the basis of gender or sexual identity is prohibited in a number of jurisdictions for a number of consumer situations. Yet the broad concept of societal fairness and the much narrower concept of actuarial fairness differ, and so within insurance markets, U.S. law has historically set insurance apart from other products in speaking to issues of fairness and discrimination (West, 2013). Within the last year, several states have enhanced their recognition of nonbinary or genderqueer identities by implementing a Gender X option on driver’s licenses. Insurance carriers are left with minimal direction on how to appropriately price this emerging class within the three goals of rate regulation. Additionally, as diversity and inclusion continue to be a strategic initiative within the insurance market, the insurance industry and its regulatory environment have to navigate carefully between the business imperatives for adequate pricing and inclusion efforts. This paper addresses the potential for unfair discrimination in some lines of business—with special focus on auto insurance—should gender-based rating be continued into the future. It also explores an immediate opportunity to enhance the insurance industry’s social compact with its insureds via recognition of the Gender X identity. Part I gives a primer on nonbinary and trans-identity followed by a brief history of the role of gender in insurance pricing, Part II discusses nonbinary, transgender, and the introduction of Gender X as an additional categorical level of the gender identify rating factor as used in insurance pricing. Part III and Part IV dive into the economic and social implications of movement in U.S. law toward more gender inclusivity.


Sign in / Sign up

Export Citation Format

Share Document