University of Sheffield

Tony O'Hagan - Academic pages - Abstracts

.

Nonparametric elicitation for heavy-tailed prior distributions

J.P. Gosling, J.E. Oakley and A. O'Hagan

University of Sheffield

Publication details: Submitted to Bayesian Analysis.


Abstract

In the context of statistical analysis, elicitation is the process of translating someoneís beliefs about some uncertain quantities into a probability distribution. The personís judgements about the quantities are usually fitted to some member of a convenient parametric family. This approach does not allow for the possibility that any number of distributions could fit the same judgements.

In this paper, elicitation of an expertís beliefs is treated as any other inference problem: the facilitator of the elicitation exercise has prior beliefs about the form of the expertís density function, the facilitator elicits judgements about the density function, and the facilitatorís beliefs about the expertís density function are updated in the light of these judgements. This paper investigates prior beliefs about an expertís density function and shows how many different types of judgement can be handled by this method.

This elicitation method begins with the belief that the expertís density will roughly have the shape of a t density. This belief is then updated through a Gaussian process model using judgements from the expert. The method gives a framework for quantifying the facilitatorís uncertainty about a density given judgements about the mean and percentiles of the expertís distribution. A property of Gaussian processes can be manipulated to include judgements about the derivatives of the density, which allows the facilitator to incorporate mode judgements and judgements on the sign of the density at any given point. The benefit of including the second type of judgement is that substantial computational time can be saved.


Return to my publications page.
Updated: 16 December 2006
Maintained by: Tony O'Hagan