Description: The objective of this paper is to present a technical exposition of the rudiments of Bayesian statistics.

Description: This paper discusses the Bayesian graduation method of Kimeldorf and Jones and develops methods for managing four technical problems in their method. It also includes a discussion of the paper. From Transactions of Society of Actuaries 1977, Vol. 29.

Description: This paper addresses various modifications of the Kimeldorf-Jones Bayesian graduation method. These enhancements entail the use of time series and eigenvalue decomposition. By augmenting the existing method one can empirically smooth the Kimeldorf-Jones technique. From Actuarial Research Clearing House 1991, VOL. 1.

Description: In this paper the ad-hoc Bayesian derivation of the Whittaker method is replaced with a formal Bayesian specification, and then it is shown that with this specification it is possible to complete the graduation without making an arbitrary selection of the smoothing parameter. From Actuarial Research Clearing House 1992, VOL. 1.

Description: The purpose of this paper is to perform conjugate Bayesian analysis on the negative binomial distribution with the goal of providing a standard by which approximate credibility formulas and approximate predictive distriubtion may be judged. From Actuarial Research Clearing House 1993, VOL. 1.

Description: This paper describes an analysis of data anomalies on computer records on 246 federally-insured mortgages. Twenty-three of the corresponding mortgages were sampled and all were found to have identical amortization plans and to have been insured under the same insurance fund. The question posed, in probabilistic terms, is the following: How many of the remaining 223 mortgages are identical in these two aspects to the 23 sampled mortgages?

Description: Presented at October 2000 Annual Meeting. This is a brief summary of a tutorial on the application of credibility theory to health insurance. The concept of full credibility and the necessary conditions for experience data to be fully credible are covered.

Description: This paper presents Bayesian models for analysis of claim lag data. The analysis of claim runoff patterns investigate the extraction of information on trend and seasonal patterns in incurred claims. This is investigated on the basis of individual group data, and also relative to a larger aggregation of pooled experience. The models are implemented based on Markov chain Monte Carlo methods. The models are applied to examples using simulated claim lag data. From Actuarial Research Clearing House, 2003, Vol. 1.

Description: Abstract from a paper that describes results from a hierarchical Bayesian statistical model which predicts the probability of Non-Acceptable Claim, an unnecessary hospital stay, using hospital discharge claim records as inputs.