Browse Research
Viewing 201 to 225 of 7679 results
Although it is an analytic construct important in its own right, a stationary population is an integral component of a life table. Using this perspective, we discuss well-known and not-so-well known equalities that are found a stationary population as well as a set of inequalities. There are two parts to the set of inequalities we discuss.
The common calculation used in developing case reserves are based on “hindsight” from a separate development test, thus they are based on data that already reflects judgment. A method is presented for estimating development factors for case reserves that strictly uses data within the standard loss development triangles, primarily the paid loss to case reserve disposed or “runoff” ratios.
The evolving definition of Advanced Analytics and the emergence of the Data Scientist
In its infancy, Actuarial Science operated at the leading edge of contemporary analytic capabilities and could be easily said to be employing “advanced analytics.” Over the past 50 years, however, relentless data and technology breakthroughs have created modern analytic capabilities that far outstrip many of our traditional actuarial pricing and reserving met
CAS E-Forum, Winter 2016 Featuring the RBC Dependencies and Calibration Subcommittee
This report starts a discussion on the interrelationships between the general levels of unemployment in the economy and the sale and persistency of insurance products. The study hypothesizes that insurance products experience reduced sales and persistency as the level of unemployment increases in a given economic environment.
Ninth Survey of Emerging Risks Risk management can be looked at in many ways—from how the volatility of an individual risk impacts profit distribution to how it threatens solvency. Emerging risks fall into the latter category. Risk managers seek out information about risks that, over a long time horizon, could have a great impact on an entity’s survival.
Large yet infrequent disruptions of electrical power can impact tens of millions of people in a single event, triggering significant economic damages, portions of which are insured. Small and frequent events are also significant in the aggregate. This article explores the role that insurance claims data can play in better defining the broader economic impacts of grid disruptions in the U.S. context.
Motivation This paper proposes a triangle-based stochastic reserving framework for parsimoniously describing insurance claims generation, reporting and settlement processes with intuitive parameters.
Method Deterministic compartmental models are explored as extensible tools to describe and project the insurance claims process using a small number of parameters, including measure of case reserve robustness.
CAS E-Forum, Fall 2016 Featuring the CAS Data & Technology Working Party Report and Independent Research
This paper provides an introduction to the use of Bayesian methods for blending prior information with a loss development pattern from a triangle. The methods build upon conjugate forms discussed in earlier literature but introduce the Generalized Dirichlet as a prior, which allows for a significant simplification in calculation.
The Cape Cod method is a commonly used technique where the a priori loss ratio is calculated as the weighted average of the chain ladder ultimate loss ratios across all years with the “used” premium as the weights. It applies the same a priori loss ratio estimate (on a trended, current rates level) across all years, without consideration for any possible changes that may have occurred.
CAS E-Forum, Summer 2016 Featuring the CAS Reserves Call Papers, Innovation Essays and Independent Research
There are two competing views on how to determine capital for an insurer whose loss liabilities extend for several time periods until settlement. The first focuses on the immediate period (usually one-year) and the second uses the runoff (until ultimate loss payment) time frame; each method will generally produce different amounts of required capital.
CAS E-Forum, Spring 2016 Featuring the CAS RBC Dependencies and Calibration Working Party Report
The underwriting elements in the NAIC Property Casualty RBC Formula (RBC Formula) are not selected to achieve a particular total safety level. We examine the historical variability in underwriting experience and measure the achieved safety level in terms of a Value at Risk (VaR). As explained in this paper, we consider a Policyholder View for measuring safety level as opposed to a Company View.
