Forecasting Product Sales with Dynamic Linear Mixture Models

Author(s):
Phillip M. Yelland and Eunice Lee
Report Number:	Date Published:	Available Formats:
TR-2003-122	March 2003	Portable Document Format (PDF) Postscript (PS) Request Hard Copy
Abstract
This paper discusses investigations undertaken at Sun Microsystems, Inc. into practical forecasting applications of the Bayesian Dynamic Linear Models described in the seminal work of West and Harrison (1997). Particular emphasis is placed on the use of class II mixture models, which use Bayesian model averaging to help accommodate model uncertainty. Keywords: Forecasting, probability, statistics, time-series.

Forecasting Product Sales with Dynamic Linear Mixture Models