Wolfram Library Archive

Courseware Demos MathSource Technical Notes
All Collections Articles Books Conference Proceedings

Bayesian Statistical Models for Financial Audits

Karl W. Heiner
Organization: State University of New York at New Paltz, USA
Anthony O'Hagan
Organization: Univeristy of Sheffield, UK
David J. Laws
Organization: Univeristy of Sheffield, UK

2003 International Mathematica Symposium
Conference location

Imperial College, London

Bayesian Statistical Models provide statisticians and auditors with a set of powerful tools for attacking financial audit problems. Some of these statistical auditing problems may be solved analytically. For these, functions for manipulating equations and standard packages make Mathematica a natural environment for work. However, most Bayesian statistical models suitable for attacking financial audit problems are not accessible in closed form. For these problems Mathematica’s symbolic programming language, efficient array handling functions, and numerical optimization routines form the base for building large-scale simulations. The beta-binomial model for estimating the proportion of transactions in error is described. Models for multilocation audits are discussed briefly, as are models for sequential multiple location audits. Why Mathematica is an important tool in this type of data analysis is shown.

*Mathematics > Probability and Statistics

Bayesian statistical models, financial audit problems, array handling functions, numerical optimization, beta-binomial model, multilocation audits, sequential multiple location audits
Related items

*Challenging the Boundaries of Symbolic Computation: Proceedings of the 5th International Mathematica Symposium   [in Books]