Skip to main content
U.S. flag

An official website of the United States government

Daniel Goodman’s empirical approach to Bayesian statistics

October 18, 2016

Bayesian statistics, in contrast to classical statistics, uses probability to represent uncertainty about the state of knowledge. Bayesian statistics has often been associated with the idea that knowledge is subjective and that a probability distribution represents a personal degree of belief. Dr. Daniel Goodman considered this viewpoint problematic for issues of public policy. He sought to ground his Bayesian approach in data, and advocated the construction of a prior as an empirical histogram of “similar” cases. In this way, the posterior distribution that results from a Bayesian analysis combined comparable previous data with case-specific current data, using Bayes’ formula. Goodman championed such a data-based approach, but he acknowledged that it was difficult in practice. If based on a true representation of our knowledge and uncertainty, Goodman argued that risk assessment and decision-making could be an exact science, despite the uncertainties. In his view, Bayesian statistics is a critical component of this science because a Bayesian analysis produces the probabilities of future outcomes. Indeed, Goodman maintained that the Bayesian machinery, following the rules of conditional probability, offered the best legitimate inference from available data. We give an example of an informative prior in a recent study of Steller sea lion spatial use patterns in Alaska.

Citation Information

Publication Year 2016
Title Daniel Goodman’s empirical approach to Bayesian statistics
DOI 10.7287/peerj.preprints.1755v1
Authors Tim Gerrodette, Eric Ward, Rebecca L. Taylor, Lisa K. Schwarz, Tomoharu Eguchi, Paul Wade, Gina Himes Boor
Publication Type Article
Publication Subtype Journal Article
Series Title PeerJ
Series Number
Index ID 70168542
Record Source USGS Publications Warehouse
USGS Organization Alaska Science Center Biology MFEB