Description: This is a letter from Jan M. Hoem to Arnold F. Shapiro about Exposed-to-risk considerations based on the Balducci assumption and other assumptions in the analysis of mortality.

Description: The objectives of this paper are to determine the extent to which Makeham's or Gompertz's laws describe recent insurance and population mortality analyze how the constants in the two laws vary with respect to some factors and propose a simple extension to Makeham's law to account for recent departures from the laws, related to difference in sex. Presented at Actuarial Research Conference at Ball State University, September 1, 1978

Description: This paper discusses a general model for calculating life contingencies including formulation of the model and net reserves and contingency reserves.

Description: Statistical properties of interest, annuity and insurance functions are examined when mortality and interest are treated as having a random component. Several special models of interest rate fluctuation are examined in detail. It is hoped that such theory will be useful in developing fluctuation reserves and premium margins. Presented at the 13th Annual Actuarial Research Conference, Ball State University, August 31st. September lst, 2nd. 1978.

Description: Recent advances in mortality modeling have enabled applicant-specific quantification of mortality risk in life insurance applicants using laboratory and physical measurements. Relative to more conventional underwriting, this approach identifies a small population of significantly under-priced policy and a much larger, cross-subsidizing group whose risk is presently overestimated. Applicants placed in the highest percentile by this scoring methodology are more than 10X as likely to die in a given period as their demographic peers, and represent approximately 5% of short-term claims. Implications for product design are substantial in both the short- and long-term, as the possibilities for adverse selection grow, and the opportunity for fully individualized pricing raises the more distant prospect of products with as many rate classes as applicants.