FORESIGHT, Issue 1

Description

Summer 2005 Issue

Special Feature: When and How Should Statistical Forecasts be Judgmentally Adjusted?

• Introduction by Feature Editor, Nada Sanders
  In her introduction to this feature section, Nada explains the challenges of judgmentally adjusting statistical forecasts, and the problems that can occur if the process is not done correctly.
• How to Integrate Management Judgment with Statistical Forecasts by Paul Goodwin
  Many of us make judgmental adjustments to statistical forecasts. But do these improve accuracy? Paul Goodwin explains when you should avoid the temptation to adjust, and shows how the accuracy of your interventions can be improved.
• Judgmental Adjustment: A Challenge for Providers and Users of Forecasts by Dilek Önkal and M. Sinan Gönül
  Both the providers and the users of forecasts appear to view judgmental adjustments as a sign of caring about the forecast. Although fine tunings are expected to increase practitioner satisfaction, adjustments may sometimes backfire and reduce forecast quality. Dilek and Sinan recommend that practitioners use caution in making adjustments, that they rely on support tools, and that they try to understand the motivations and expectations behind the adjustment process.
  • Commentary by Anthony Lee
• Relative Merits of Different Ways of Combining Judgment with Statistical Forecasts by Nigel Harvey
  Instead of using your judgment to combine judgmental and statistical forecasts, just average them. Also, using your judgment to adjust a statistical forecast is better than modifying your judgmental forecast on the basis of a statistical one. Judgmental adjustment is likely to be more successful if you are a domain expert, if you use decomposition, if more than one person estimates the size of the required adjustment, if you have knowledge not used in the statistical forecast, and if you keep records that allow you to see whether your past adjustments have been too large or too small.
  • Commentary by Lucy Kjolso

Articles

• 1. The Forecasting Canon: Nine Generalizations to Improve Forecast Accuracy by J. Scott Armstrong
     Using findings from empirically-based comparisons, Scott develops nine generalizations that can improve forecast accuracy. He finds that these are often ignored by organizations, so that attention to them offers substantial opportunities for gain. In this paper, Scott offers recommendations on how to structure a forecasting problem, how to tap managers’ knowledge, and how to select appropriate forecasting methods.
  2. Intermittent and Lumpy Demand: A Forecasting Challenge by John Boylan
     Slow items with intermittent and lumpy demand patterns may seem unimportant, but they can make up a substantial part of an organization’s inventory. They are difficult to forecast, and some of the most popular forecasting methods are unsuitable or in need of improvement. In this paper, John describes the principal methods for forecasting intermittent and lumpy demands and shows how to take advantage of important recent advances in this field.
  3. Case Study: Integrating Consumer Demand to Improve Shipments Forecasts by Charles W. Chase, Jr.
     The consumer packaged-goods industry (CPG) gives high priority to linking forecasts of consumer demand to shipments forecasts, in order to capture the impact of marketing activities on factory shipments. In this case study of a CPG company, Charlie Chase shows how to consider marketing and replenishment strategies jointly rather than individually.
  4. How to Evaluate the Forecasting Ability of Demand-Planning Software by Jim Hoover
     In this column, Jim examines the functionality and implementation of demand-planning software, a market that has received very little scrutiny to date in forecasting books and journals.
     Book Review by Nada Sanders
     Demand Management Best Practices by Colleen Crum with George Palmatier