Description: This paper analyzes the U. S. Treasury monthly interest rates from 1953 to 1992, and the daily rates from 1981 to 1992 for various forms of chaotic behavior. The primary analysis was in determining the Burst Exponent by the use of Rescaled Range Analysis. This paper also discusses other possible uses of Fractional Brownian Motion in Finance Theory, Macroeconomics, and Actuarial Science. From Actuarial Research Clearing House 1994 VOL. 1.

Description: This article list descriptions of the sessions that are either sponsored or co–sponsored by the Education and Research Section.

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.

Description: In this presentation, the authors develop a stochastic generator for the generation of scenarios of the S & P 500 index, dividend yield, consumer price index, and U.S. Treasury yields. They first create a set of 'stylized facts' for these series. These in-sample statistical models are themselves not suitable for generation of scenarios for decision making, but instead are additional stylized facts that assist in model development. The 'best' statistical model according to standard statistical model selection criteria can easily lead to a model that is highly unsuitable for generation of scenarios for decision making. The authors develop a stochastic generator that is suitable for decision making under uncertainty. From Actuarial Research Clearing House 1997, Vol. 1.

Description: This article recaps session 19 "Estimating Mortality Risk Using Predictive Modeling".

Description: Abstract for the 2008 ERM Monograph paper, “Estimating Mortality Risk Using Predictive Modeling”

Description: This 1997 presentation from the 32nd Actuarial Research Conference develops a stochastic generator for the generation of scenarios of the S&P 500 index, dividend yield, consumer price index, and U.S. Treasury yields. The authors first create a set of 'stylized facts' for these series. They estimate statistical models for these series. These in-sample statistical models are themselves not suitable for generation of scenarios for decision making, but instead are additional 'stylized facts' that assist in model development. The 'best' statistical model according to standard statistical model selection criteria can easily lead to a model that is highly unsuitable for generation of scenarios for decision making. The authors develop a stochastic generator that is suitable for decision making under uncertainty.

Description: This is the abstract of a paper in which the authors develop sixteen separate mortality models based on the current SOA mortality data being used for the preferred life study. The models are constructed using projection pursuit regression. The sixteen models arise from all combinations of sex, smoker status, select and ultimate status, and per life vs. per dollar mortality rates. In general the models confirm the following four facts: 1 Mortality rates increase as attained age increases. 2 Mortality rates increase as duration increases. 3 Mortality rates decrease as Face Amount Bands increase. 4. Mortality rates decrease if preferred underwriting is present.

Description: These are the slides that go with the paper 'Mortality Modeling using Projection Pursuit Regression.' In that paper, the authors develop sixteen separate mortality models based on the current SOA mortality data being used for the preferred life study. The models are constructed using projection pursuit regression. The sixteen models arise from all combinations of sex, smoker status, select and ultimate status, and per life vs. per dollar mortality rates. In general the models confirm the following four facts: 1 Mortality rates increase as attained age increases. 2 Mortality rates increase as duration increases. 3 Mortality rates decrease as Face Amount Bands increase. 4. Mortality rates decrease if preferred underwriting is present.