Bayesian ratemaking procedure of crop insurance contracts with skewed distribution

Over the years, crop insurance programs became the focus of agricultural policy in the USA, Spain, Mexico, and more recently in Brazil. Given the increasing interest in insurance, accurate calculation of the premium rate is of great importance. We address the crop-yield distribution issue and its implications in pricing an insurance contract considering the dynamic structure of the data and incorporating the spatial correlation in the Hierarchical Bayesian framework. Results show that empirical (insurers) rates are higher in low risk areas and lower in high risk areas. Such methodological improvement is primarily important in situations of limited data.

Parametric and nonparametric statistical modelling of crop yield: implications for pricing crop insurance contracts

This article considers alternative methods to calculate the fair premium rate of crop insurance contracts based on county yields. The premium rate was calculated using parametric and nonparametric approaches to estimate the conditional agricultural yield density. These methods were applied to a data set of county yield provided by the Statistical and Geography Brazilian Institute (IBGE), for the period of 1990 through 2002, for soybean, corn and wheat, in the State of Paran. In this article, we propose methodological alternatives to pricing crop insurance contracts resulting in more accurate premium rates in a situation of limited data.

Matching prescription claims with medication data for nursing home residents: Implications for prescriber feedback, drug utilisation studies and selection of prescription claims database

Medication data retrieved from Australian Repatriation Pharmaceutical Benefits Scheme (RPBS) claims for 44 veterans residing in nursing homes and Pharmaceutical Benefits Scheme (PBS) claims for 898 nursing home residents were compared with medication data from nursing home records to determine the optimal time interval for retrieving claims data and its validity. Optimal matching was achieved using 12 weeks of RPBS claims data, with 60% of medications in the RPBS claims located in nursing home administration records, and 78% of medications administered to nursing home residents identified in RPBS claims. In comparison, 48% of medications administered to nursing home residents could be found in 12 weeks of PBS data, and 56% of medications present in PBS claims could be matched with nursing home administration records. RPBS claims data was superior to PBS, due to the larger number of scheduled items available to veterans and the veteran's file number, which acts as a unique identifier. These findings should be taken into account when using prescription claims data for medication histories, prescriber feedback, drug utilisation, intervention or epidemiological studies. (C) 2001 Elsevier Science Inc. All rights reserved.

Whole life health insurance

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para obtenção do Grau de mestre em Matemática e Aplicações

Exploraty Multivariate Statistical Methods Applied to Pharmaceutical Industry CRM Data

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

Statistics as a basis for discrimination in the insurance business

The European Court of Justice has held that as from 21 December 2012 insurers may no longer charge men and women differently on the basis of scientific evidence that is statistically linked to their sex, effectively prohibiting the use of sex as a factor in the calculation of premiums and benefits for the purposes of insurance and related financial services throughout the European Union. This ruling marks a sharp turn away from the traditional view that insurers should be allowed to apply just about any risk assessment criterion, so long as it is sustained by the findings of actuarial science. The naïveté behind the assumption that insurers’ recourse to statistical data and probabilistic analysis, given their scientific nature, would suffice to keep them out of harm’s way was exposed. In this article I look at the flaws of this assumption and question whether this judicial decision, whilst constituting a most welcome landmark in the pursuit of equality between men and women, has nonetheless gone too far by saying too little on the million dollar question of what separates admissible criteria of differentiation from inadmissible forms of discrimination.

Fiscal insurance and debt management in OECD economies

Assuming the role of debt management is to provide hedging against fiscal shocks we consider three questions: i) what indicators can be used to assess the performance of debt management? ii) how well have historical debt management policies performed? and iii) how is that performance affected by variations in debt issuance? We consider these questions using OECD data on the market value of government debt between 1970 and 2000. Motivated by both the optimal taxation literature and broad considerations of debt stability we propose a range of performance indicators for debt management. We evaluate these using Monte Carlo analysis and find that those based on the relative persistence of debt perform best. Calculating these measures for OECD data provides only limited evidence that debt management has helped insulate policy against unexpected fiscal shocks. We also find that the degree of fiscal insurance achieved is not well connected to cross country variations in debt issuance patterns. Given the limited volatility observed in the yield curve the relatively small dispersion of debt management practices across countries makes little difference to the realised degree of fiscal insurance.

Changes in the demand for private medical insurance following a shift in tax incentives

The 1998 Spanish reform of the Personal Income Tax eliminated the 15% deduction for private medical expenditures including payments on private health insurance (PHI) policies. To avoid an undesirable increase in the demand for publicly funded health care, tax incentives to buy PHI were not completely removed but basically shifted from individual to group employer-paid policies. In a unique fiscal experiment, at the same time that the tax relief for individually purchased policies was abolished, the government provided for tax allowances on policies taken out through employment. Using a bivariate probit model on data from National Health Surveys, we estimate the impact of said reform on the demand for PHI and the changes occurred within it. Our findings suggest that the total probability of buying PHI was not significantly affected. Indeed, the fall in the demand for individual policies (by 10% between 1997 and 2001) was offset by an increase in the demand for group employer-paid ones, so that the overall size of the market remained virtually unchanged. We also briefly discuss the welfare effects on the state budget, the industry and society at large.

An Introduction to parametric and non-parametric models for bivariate positive insurance claim severity distributions

We present a real data set of claims amounts where costs related to damage are recorded separately from those related to medical expenses. Only claims with positive costs are considered here. Two approaches to density estimation are presented: a classical parametric and a semi-parametric method, based on transformation kernel density estimation. We explore the data set with standard univariate methods. We also propose ways to select the bandwidth and transformation parameters in the univariate case based on Bayesian methods. We indicate how to compare the results of alternative methods both looking at the shape of the overall density domain and exploring the density estimates in the right tail.

Mixture of bivariate Poisson regression models with an application to insurance

In a recent paper Bermúdez [2009] used bivariate Poisson regression models for ratemaking in car insurance, and included zero-inflated models to account for the excess of zeros and the overdispersion in the data set. In the present paper, we revisit this model in order to consider alternatives. We propose a 2-finite mixture of bivariate Poisson regression models to demonstrate that the overdispersion in the data requires more structure if it is to be taken into account, and that a simple zero-inflated bivariate Poisson model does not suffice. At the same time, we show that a finite mixture of bivariate Poisson regression models embraces zero-inflated bivariate Poisson regression models as a special case. Additionally, we describe a model in which the mixing proportions are dependent on covariates when modelling the way in which each individual belongs to a separate cluster. Finally, an EM algorithm is provided in order to ensure the models’ ease-of-fit. These models are applied to the same automobile insurance claims data set as used in Bermúdez [2009] and it is shown that the modelling of the data set can be improved considerably.

A logistic regression approach to estimating customer profit loss due to lapses in insurance

This article focuses on business risk management in the insurance industry. A methodology for estimating the profit loss caused by each customer in the portfolio due to policy cancellation is proposed. Using data from a European insurance company, customer behaviour over time is analyzed in order to estimate the probability of policy cancelation and the resulting potential profit loss due to cancellation. Customers may have up to two different lines of business contracts: motor insurance and other diverse insurance (such as, home contents, life or accident insurance). Implications for understanding customer cancellation behaviour as the core of business risk management are outlined.

How to use the standard model with own data?

In this work discuss the use of the standard model for the calculation of the solvency capital requirement (SCR) when the company aims to use the specific parameters of the model on the basis of the experience of its portfolio. In particular, this analysis focuses on the formula presented in the latest quantitative impact study (2010 CEIOPS) for non-life underwriting premium and reserve risk. One of the keys of the standard model for premium and reserves risk is the correlation matrix between lines of business. In this work we present how the correlation matrix between lines of business could be estimated from a quantitative perspective, as well as the possibility of using a credibility model for the estimation of the matrix of correlation between lines of business that merge qualitative and quantitative perspective.

Review of Measures to Reduce Costs in the Private Health Insurance Market 2014

5.11.2014 This report was prepared independently by Mr McLoughlin with the support of the health insurers, and the Health Insurance Authority, for consideration by the Minister for Health and the insurers.  All parties were very conscious of the importance of respecting competition law when dealing with issues such as prices and costs. The work of the Group has been conducted in two phases, with the first phase report published on 26 December 2013. The Phase 1 report sets out the context, establishment, membership and terms of reference for both phases of the Groups work.  The report also outlines the legislative provisions for private health insurance in Ireland, the objectives of both phases of the review and the approach and methodology followed. Phase 2 of the process focused on the compilation and analysis by the Health Insurance Authority (HIA) of claims data to assess the cost drivers for health insurance, the effects of medical technology and innovations on costs, and claims processing issues.The report and submissions from relevant stakeholders which were examined and considered under the Phase 2 Review can be downloaded below. Download the Review of Measures to Reduce Costs in the Private Health Insurance Market 2014 -  Independent Report to the Minister for Health and Health Insurance Council here. Submissions received HSE Submission to Pat McLoughlin, Chair of Review Group IHAI submission 11 April 2014 IHCA submission to Chair 1 May 2014 Insurance Ireland submission Society of Actuaries in Ireland submission St. Patricks Mental Health Services submission April 2014 St John of Gods Submission        

Use of variability in national and regional data to estimate the prevalence of lymphangioleiomyomatosis.

Background: Understanding the true prevalence of lymphangioleiomyomatosis (LAM) is important in estimating disease burden and targeting specific interventions. As with all rare diseases, obtaining reliable epidemiological data is difficult and requires innovative approaches.Aim: To determine the prevalence and incidence of LAM using data from patient organizations in seven countries, and to use the extent to which the prevalence of LAM varies regionally and nationally to determine whether prevalence estimates are related to health-care provision.Methods: Numbers of women with LAM were obtained from patient groups and national databases from seven countries (n = 1001). Prevalence was calculated for regions within countries using female population figures from census data. Incidence estimates were calculated for the USA, UK and Switzerland. Regional variation in prevalence and changes in incidence over time were analysed using Poisson regression and linear regression.Results: Prevalence of LAM in the seven countries ranged from 3.4 to 7.8/million women with significant variation, both between countries and between states in the USA. This variation did not relate to the number of pulmonary specialists in the region nor the percentage of population with health insurance, but suggests a large number of patients remain undiagnosed. The incidence of LAM from 2004 to 2008 ranged from 0.23 to 0.31/million women/per year in the USA, UK and Switzerland.Conclusions: Using this method, we have found that the prevalence of LAM is higher than that previously recorded and that many patients with LAM are undiagnosed.

Cross-coding from the Swiss Disability Insurance listing to the ICIDH.

It has been suggested that converting, via a process of cross-coding, the listing used by the Swiss Disability Insurance (SDI) for their statistics into codes of the International Classification of Impairments, Disabilities, and Handicaps (ICIDH) would improve the quality and international comparability of disability statistics for Switzerland. Using two different methods we tested the feasibility of this cross-coding on a consecutive sample of 204 insured persons, examined at one of the medical observation centres of the SDI. Cross-coding is impossible, for all practical purposes, in a proportion varying between 30% and 100%, depending on the method of cross-coding, the level of disablement and the required quality of the resulting codes. Failure is due to lack of validity of the SDI codes: diseases are poorly described, consequences of diseases (disability and handicap, including loss of earning capacity), insufficiently described or not at all. Assessment of disability and handicap would provide necessary information for the SDI. It is concluded that the SDI should promote the use of the ICIDH in Switzerland, especially among medical practitioners whose assessment of work capacity is the key element in the decision to award benefits or propose rehabilitation.