View Table of Contents »


Spring 2008 Issue

Special Feature: Prediction Markets For Pharmaceutical Forecasting And Beyond

  • A Guide to Practical Adoption in the Pharmaceutical Industry by Carol Gebert
    While prediction markets have attracted considerable attention as potential forecasting systems, very little has been written about the challenges of implementing them. Carol addresses this need here, making her case that successful implementation requires predictions worth making, an active community of participants, and real reward incentives for participating.
  • Defining Events and Motivating Participation by Andreas Graefe
    Andreas extends Carol’s discussion of the hurdles to implementation of prediction markets with his thoughts on defining the prediction event and the often delicate issue of the types of incentives needed to motivate trader participation in the market.
  • A Primer on Prediction Markets by Joe Miles
    prediction market is a virtual market whose purpose is forecasting something. It relies on people trading futures contracts whose value is tied to a particular event. Once trading has started, the market price of futures contracts will change to incorporate information that traders have about the coming event.


    1. Book Review by Roy Batchelor
      Competing on Analytics: The New Science of Winning by Thomas H. Davenport and Jeanne G. Harris
    2. Hot New Research Column: Predicting the Demand for New Products by Paul Goodwin
      Predicting demand for new products poses special problems for forecasters. By definition, a new product has little or no demand history. Unless the demand histories of similar existing products are available and considered relevant, the use of statistical approaches to detect and extrapolate past demand patterns (like exponential smoothing or ARIMA models) are usually ruled out.
    3. The Value of Information Sharing in the Retail Supply Chain: Two Case Studies by Tonya Boone and Ram Ganeshan
      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. A frequent consequence is the bullwhip effect . Using two case studies, Tonya and Ram show how information sharing – both within the company’s boundaries and with external partners – can mitigate the bullwhip effect and reduce supplychain costs.
    4. Innovations in Sales Forecasting for Large-Scale Retailers by Bruce Andrews, James Bennett, Lindsey Howe, Brooks Newkirk, and Joseph Ogrodowczyk
      Working as a team for the Center for Business and Economic Research at the University of Southern Maine, Bruce, James, Lindsey, Brooks and Joseph created a forecasting system for a large retail chain. Their base model uses the ARIMA methodology of Box and Jenkins, but the team has extended ARIMA to deal with the significant challenges of forecasting weekly sales at store and departmental levels. This article offers a case study in the modeling of weekly retail sales and a comprehensive overview of the forecasting hurdles that retailers must address. The modeling innovations noted here should provide retail forecasters with many good leads to pursue.
    5. Monte Carlo Simulation/Risk Analysis on a Spreadsheet: Review of Three Software Packages by Sam Sugiyama
      Analysts must deal with uncertainty in virtually every forecast they generate. One approach to assessment of uncertainty is Monte Carlo simulation. In an earlier Foresight article (Sugiyama, 2007), Sam provided an overview and illustration of how Monte Carlo simulation can enrich the forecasting process. Here he reviews the three major software offerings on the market today.
    6. “Been There, Done That”: Perils, Pitfalls, and Promises of Long-Term Projections by Ira Sohn
      Twenty-five years ago, a team led by Nobel laureate Wassily Leontief applied the United Nations World Input-Output Model to make forecasts of world fuel and mineral resources for the year 2000 and beyond. Ira Sohn, a member of that team, reflects on his experience with this long-term forecasting endeavor. Noting that long-term forecasting is fraught with uncertainties, Ira analyzes the sources of the forecasting errors made and suggests ways to improve long-term modeling.
Go to Top