Effectively Combining Judgment with Statistical Forecasts
Judgmental and statistical forecasts each have their own strengths and weaknesses and can bring different information to the forecasting process. This session will discuss different ways of combining judgment and statistical forecasts and the advantages and disadvantages of each approach, factoring in the inherent biases of each method. Dr. Sanders will also discuss principles that have been developed for deciding when and how to use judgment in adjusting statistical forecasts, based on extensive research and the real-world experience of prominent forecasting researchers.

Dr. Nada R. Sanders
Professor
Neeley School of Business, TCU

Forecasting the Presidential Election
This session provides an overview of the methods commonly used to forecast presidential elections and assesses their accuracy in predicting the outcome of recent elections. Using the most historically accurate techniques, Dr. Jones will present forecasts of both the popular vote and electoral vote for 2008. As will become apparent, the most accurate forecasts of presidential elections are usually generated by techniques that are relatively simple and straightforward. In particular, the gain in accuracy from combining forecasts—created by averaging the results of various polls and other methods—will be demonstrated. Internet-based data sources will be described so that participants can make and update their own forecasts of the election as the fall campaign progresses.

Dr. Randall Jones
Professor, Political Science
University of Central Oklahoma

Organizational Politics of Forecasting: Strategies for Overcoming Bias in the Forecast Process
A technically sound forecast goes nowhere if it is not accepted by those in power, whether they are politicians or upper management. Often the forecast gets manipulated to satisfy political ends or meet targets and plans. This presentation tackles some tough issues that organizations face, such as:

 How can we minimize organizational biases that impact forecasting?
 How can we improve collaboration among players with conflicting interests?
 How can we achieve buy-in or acceptance of the forecast by those in power?
 What’s the best organizational design for a neutral forecast process?

This session will discuss simple strategies to overcome these problems, allowing the technically-sound forecast to prevail over politics.

Elaine Deschamps, Ph. D.
Senior Fiscal Analyst
Washington State Senate

How Sales Forecast “Gaming” Hurts Supply Chains
The term “gaming” refers to dysfunctional behavior which may appear to be helpful to the organization although it actually has a negative effect. For example, a sales manager may set an unrealistically low forecast so that his team exceeds its targets, but this “low ball” forecast results in a product shortage at a critical time. In this session, Dr. Mello will share research that identifies specific sales forecasting gaming activities, and demonstrate the negative impact such activities can have on supply chains within and between organizations. He will also discuss ways to spot gaming in the sales forecasting process and propose methods for eliminating these counterproductive behaviors.

Dr. John E. Mello
Assistant Professor of Marketing
Arkansas State University

Forecasting and Futuring to See Over the Horizon
Traditional forecasting methods extrapolate past relationships into the future, on the assumption that past relationships will at most change incrementally. For the typical forecast horizon of two years or less, low error may be achieved for a number of rolling forecasts; however, repeated continually over a ten-year period, or even a five-year period for many products, large errors are nearly certain to occur. In today’s information age, knowledge is acquired and diffused across fields at increasing speeds, giving birth ever more quickly to futures very different from the past and present. Instead of incremental change, the forecaster increasingly is faced with unexpected structural breaks or regime shifts which can produce forecasts that are disastrously wrong about the nature and pace of change—with errors outside of any confidence range based on the assumption of stable relationships. Traditional forecasting methods need to be complemented with futuring, which at its core is the anticipation of future shifts. Doing it well requires a broad, multi-disciplinary point of view and paying special attention to how the convergence or fusion of seemingly diverse changes may shift what will be produced; who will do it and how; and how customers will buy it and use it. Dr. Pearson will discuss what some traditional forecasts five to fifteen years ahead offer as benchmarks, and then describe how futuring and its methodology may improve the accuracy of short-term as well as long-term forecasting processes.

Dr. Roy L. Pearson
Chancellor Professor Emeritus
College of William and Mary

Panel of Experts: Meeting Forecasting Challenges
This session gives you the opportunity to ask direct questions of a panel of internationally-recognized experts in business forecasting. The questions may address any aspect of forecasting, including the management of the forecasting function, organizational impediments to good forecasting practice, establishment of performance targets and methods and software appropriate for the company’s products. Here is a sampling of questions raised at the last Forecasting Summit:

 How do you effectively set up a forecasting team? Where is the best place for the forecasting function to reside – marketing, finance, operations?
 Our company is trying to work from a single forecast for all groups – marketing, sales, finance and supply-chain. Is this realistic? What are the pitfalls?
 Upper management supplies a sales target. How do you forecast to avoid a “self-fulfilling prophesy”?
 How can a forecaster avoid being blamed for the failure of product sales to meet plan?
 How do you forecast sales in a company whose promotional campaigns do not match the work months?
 How can a company establish a target for forecast accuracy?
 How can a company steer between the pitfalls of (a) overly frequent forecast updates, thus introducing excessive variability and (b) failure to make timely updates, perhaps missing the market?

Moderator:
Dr. Len Tashman
Professor Emeritus
University of Vermont (UVM)

Harnessing Collective Knowledge—Using Prediction Markets to Improve Your Forecasts
Prediction markets—which allow participants to bet on the outcome of an event — have been found to be valuable in election forecasting. In response to this success, an increasing number of companies are harnessing prediction markets to forecast critical business outcomes such as sales volumes, product launch dates and the success of new products. This session introduces the concept of prediction markets, provides evidence from the literature about their performance for business forecasting and discusses implementation challenges. Mr. Graefe will also report on recent research in which he has analyzed the ability of prediction markets to aggregate information from small groups by comparing their forecasts to those derived through traditional meetings and via the Delphi Technique.

Andreas Graefe
Research Associate
Institute for Technology Assessment and Systems Analysis

The Value of Information Sharing in the Retail Supply Chain: Two Case Studies
Retail supply chains are complex, with each company in the chain having multiple echelons of distribution. Forecasting and requirements planning are further challenged by managers’ reliance on “local” rather than chain-wide retail demand to make key operational decisions. Using two case studies, this session illustrates how information sharing—both within the company’s boundaries and with external partners—can improve operational efficiency and reduce supply-chain costs.

Ram Ganeshan
Associate Professor
Mason School of Business

  Business Forecast Systems in cooperation with the International Institute of Forecasters