Posterior distributions for rare events in multivariate categorical data

Authors:
Douglas H. Jones
Affiliations:
Management Science and Information Systems, Rutgers Business School-Newark and New Brunswick, Piscataway, NJ
Venue:
ISP'07 Proceedings of the 6th WSEAS international conference on Information security and privacy
Year:
2007

Citing 3
Cited 0

A computational algorithm for handling the special uniques problem

International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems
A New Algorithm for Finding Minimal Sample Uniques for Use in Statistical Disclosure Assessment

ICDM '05 Proceedings of the Fifth IEEE International Conference on Data Mining
Using discrete multivariate mcmc bayesian methods for change detection and disclosure control

Using discrete multivariate mcmc bayesian methods for change detection and disclosure control

Quantified Score

Hi-index	0.00

Visualization

Abstract

This study proposes a method to estimate the posterior distribution of multidimensional categorical data. This methodology enables Bayesian analysis of rare events by borrowing strength from a large database. Once the posterior distributions are profiled, further analysis can be performed and/or decisions made about importance of the occurrence of a particular rare event. For example, the occurrence of a rare event can signal an unusual or undesirable activity in a supply chain and lead to instability in vendors or suppliers and other chain components, possibly leading to the failure of the entire supply chain. Some supply chains are critical for a stable economy and national security, thus early and efficient detection of disruptions of these supply chains are essential.