12. An Evaluation of Croston’s Method under Non-Stationary Demand Behavior

—Matthew Lindsey, Stephen F. Austin State University

—Robert Pavur, University of North Texas

Abstract

Intermittent-demand data—that is, series with many periods in which no demand occurs—is problematic for accurately modeling and predicting future demand. Croston’s method is generally viewed as the “standard” method for intermittent-demand forecasting. Despite almost two decades of proposed modifications and assessments of drawbacks of this procedure, clear guidelines as to its performance under a wide variety of modeling assumptions are lacking. In this study, an assessment of Croston’s method is conducted in a simulation study to evaluate its comparative performance ...

Get The Best Thinking in Business Analytics from the Decision Sciences Institute now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.