Description: This paper demonstrates the calculation of the moments of the distribution of aggregate life insurance claims from seriatim inforce data, assuming that each record represents an independent Bernoulli trial with its own known probability of a claim, q. Continuous distributions matching the first three or four moments are identified.

Description: This article is a response to the article Current U.S. Ordinary Life Mortality by Doug Doll and Mike Jones. The author offers his interpretation of the mortality improvements observed between 1980-85 and 1983-88.

Description: This paper investigates the use of models of dependent mortality for determining annuity values. We discuss a broad class of parametric models using a bivariate survivorship function called a copula. Using data from a large insurance company to illustrate our methods, we calculate maximum likelihood estimates to calibrate the rnodel.

Description: This paper develops the binomial version of the Lee-Carter model and provides a comparative study of simulation strategies for assessing risk in mortality rate predictions and associated estimates of life expectancy and annuity values in both period and cohort frameworks.

Description: We propose an age-shift model to modify the LC model and deal with the problem of parameters. The proposed method attains smaller estimation errors with respect to mean absolute percentage error or mean square error. We also apply the proposed age-shift model to the mortality rates of these four countries to evaluate the longevity risk in annuity products by measuring life expectancy.

Description: Mortality in the U.S. states at ages 80 and older and over years 1959–2011 analyzed and compared with mortality levels in 13 high-longevity countries.

Description: We ask whether we can find support for the previously reported upper limit of the force of mortality of 0.7, i.e. about the probability of tossing a fair coin.

Description: These are the discussant comments for the concurrent session 4A: Longevity in the Public Eye from the January 4-6, 2017 Living to 100 Symposium in Orlando.

Description: Presented at Living to 100 symposium, January 2002. This is a report of a study in which the Cox proportional hazard model, a widely used survival analysis tool, was applied to profiling the mortality of an insured advanced-age population. The data was derived from 14 US insurers. This data represents policies issued between 01/01/1997 and 12/31/2000 to policyholders of age 60 or older. The major study results include measures of mortality, variations by selected risk factors, estimated mortality experience of the underlying population, and a comparison of mortality experience estimated by direct calculation, the Society of Actuaries’ SOA 1990-95 mortality study, and our extended Cox modeling technique.

Description: As a life reinsurer, understanding mortality risk is mission critical. Superior models of mortality risk lead to more competitive pricing in profitable market segments and more efficient use of capital. Surprisingly, modern statistical tools such as predictive modeling have not previously been used to model mortality data to gain this edge. In this paper, we present the use of a predictive modeling technique called projection pursuit regression PPR to model mortality data. We first describe the process of creating a PPR model using inter-company mortality data collected by the Individual Life Experience Committee of the Society of Actuaries, and present the resulting model. We then discuss the process of using this model as a starting point for creating a more detailed PPR model of internal company mortality data.