scispace - formally typeset
Search or ask a question

Showing papers by "Barry C. Arnold published in 1996"


Book ChapterDOI
01 Jan 1996
TL;DR: In this article, the maximal value of an attribute in a given generation and in the entire process is studied, and the associated limiting distributions are shown to satisfy a well known functional equation.
Abstract: If each individual in a branching process has an associated measured attribute (e.g. height), it is natural to study the maximal value of the attribute in a given generation and in the entire process. The associated limiting distributions are shown to satisfy a well known functional equation.

22 citations


Journal ArticleDOI
TL;DR: In this paper, a k-dimensional density function is determined by certain combinations of marginal and conditional densities, and appropriate compatibility conditions are described as are sufficient conditions for uniqueness of the resulting density which must, in general, be checked for integrability.

19 citations


Book ChapterDOI
01 Jan 1996
TL;DR: In this paper, a characterization of such distributions is provided and related estimation problems are discussed, where the authors focus on models in which these conditional survival functions are of the proportional hazards form.
Abstract: Bivariate survival models can sometimes be characterized in terms of conditional survival functions of the form P(X > x|Y > y) and P(Y > y|X > x). Attention is focussed on models in which these conditional survival functions are of the proportional hazards form. A characterization of such distributions is provided and related estimation problems are discussed.

10 citations


Journal ArticleDOI
TL;DR: In this paper, the authors select a family of prior distributions to combine with a given likelihood, the prime consideration is that the resulting posteriors should be members of tractable families of distributions.
Abstract: When selecting a family of prior distributions to combine with a given likelihood, the prime consideration is that the resulting posteriors should be members of tractable families of distributions. Assuming a likelihood belonging to an exponential family we identify the family of priors which will lead to convenient posteriors in the sense that they belong to a specified exponential family.

10 citations


Journal ArticleDOI
TL;DR: Functional equations enable us to arrive at a simple consensus model for fatigue life of longitudinal elements based on engineering principles and attention is given to the desirability of deriving models exhibiting plausible asymptotic independence.
Abstract: Functional equations enable us to arrive at a simple consensus model for fatigue life of longitudinal elements based on engineering principles. The model is then applied to two representative data sets. By the application of proportional hazards techniques and subsequent likelihood analysis, simple parsimonious Weibull models are derived. Inter alia attention is given to the desirability of deriving models exhibiting plausible asymptotic independence.

5 citations


Journal ArticleDOI
TL;DR: In this paper, the authors studied the time between consecutive gas release events at Hanford nuclear waste storage tanks and established time windows in which waste tank research and maintenance activities can be safely performed.
Abstract: Certain radioactive waste storage tanks at the United States Department of Energy Hanford facilities continuously generate gases as a result of radiolysis and chemical reactions. The congealed sludge in these tanks traps the gases beneath it and causes the level of the waste within the tanks to rise. The waste level continues to rise until the sludge loses its buoyancy and rolls over, changing places with fluid on top. During a rollover, the trapped gases are released, resulting in a sudden drop in the waste level. This is known as a gas release event (GRE). After a GRE, the waste re-congeals and gas again accumulates, leading to another GRE. We are interested in the time between consecutive GREs. Understanding the probabilistic behaviour of the time between consecutive GREs is important because the hydrogen and nitrous oxide gases released during a GRE are flammable and the ammonia that is released is a health risk. From a safety perspective, activity around such waste tanks should be halted when a GRE is imminent. With a credible probability model for the time between consecutive GREs, we can establish time windows in which waste tank research and maintenance activities can be safely performed. We discuss the application of non-linear time series models to this problem.