Description: Provides white vs. black mortality rates based upon Social Security data for years 1984 - 1988, and separate tables based upon data from 'Health United States' for years 1983 - 1987 and data from 'Vital Statistics of the United States' for years 1983 - 1987.

Description: Some recent studies have shown a 'crossover' of mortality rates, with female rates exceeding male rates beyond some age around 100. The purpose of this paper is to develop some statistical tests for the significance of observed crossovers. From Transactions of Society of Actuaries 1986, Vol. 38.

Description: This paper examines the effects of selection procedures used in epidemiologic studies and suggests that similar procedures could be employed to select standard ordinary insurance risks at the older ages. The experience in this paper was developed from a prospective epidemiologic investigation launched by the American Cancer Society in the fall of 1959 to learn how personal characteristics, living styles, and environmental conditions influence health and longevity.

Description: This paper was motivated in part by a desire to use, in a variety of classroom situations, some of the readily available statistics on mortality and related factors. A second aim was to illustrate certain statistical ideas, such as the relative merits of means and medians. However, the subject of racial differentials in mortality rates is of considerable intrinsic interest, which the analysis presented here only begins to address.

Description: In this paper, Bivariate Schuette graduation is applied to mortality rates among black male retired workers [Social Security data] at ages 70 to 83. Graduated values for various choices of the parameters are compared. The Schuette method is also adapted to calculate regression quantiles, providing upper and lower envelopes for the graduated values.

Description: Australia has conducted ten national censuses since World War II: in 1947, 1954, 1961 and at five-year intervals through 1996. Life tables have been prepared in connection with each of these censuses except the most recent one, for which data is still being collected.

Description: By disinformation we mean the inclusion of data that should not be considered in a study. In this paper we discuss a case in which the presence of disinformation was clear, but it was not possible to determine with certainly which records were extraneous.

Description: The adjustment coefficient R plays an important role in mathematical ruin theory this paper presents several different methods for estimating the variance of R. Examples are given in which the variance is so high that the estimate of R must be considered essentially worthless and other means are sought for estimating ruin probabilities. From the Actuarial Research Clearing House ARCH 1990 Vol. 1.

Description: Discussion of paper presented at Living to 100 and Beyond symposium. Discusses paper by Guo and Wang on the use of data mining to determine the key factors affecting mortality rates at older ages.

Description: Two-dimensional data can be smoothed by minimizing a combination of the sum of weighted absolute deviations from initial estimates and the sums of absolute values of differences in the two directions. This paper discusses the techniques of parametric programming which make it feasible to explore the different results obtained by varying the emphasis given to fit versus smoothness. From Actuarial Research Clearing House 1994, VOL. 1.