A study of comparative forecasting.

Abstract

During the past two decades, there has been an increasing number of comparative forecasting studies. The objective of these studies is to compare different forecasting methodologies with the hope of finding the best methodology. These studies have led to conflicting reports and controversies.

This dissertation examines almost all published comparative studies and delineates a list of fallacies occurring in comparative forecasting studies. These fallacies most commonly give rise to the existing controversies. Since the controversies in forecasting stem from comparisons of the various approaches, a brief synopsis of the most currently employed univariate and multivariate methodologies are presented.