Prerelease Demand Forecasting for Motion Pictures Using Functional Shape Analysis of Virtual Stock Markets

Prerelease demand forecasting is one of the most crucial yet difficult tasks facing marketers in the $60 billion motion picture industry. We propose functional shape analysis (FSA) of virtual stock markets (VSMs) to address this long-standing challenge. In VSMs, prices of a movie's stock reflect the dynamic demand expectations prior to the movie's release. Using FSA, we identify a small number of distinguishing shapes, e.g., the last-moment velocity spurt, that carry information about a movie's future demand and produce early and accurate prerelease forecasts. We find that although forecasting errors from the existing methods, e. g., those that rely on movie features, can be as high as 90.87%, our approach results in an error of only 4.73%. Because demand forecasting is especially useful for managerial decision making when provided long before a movie's release, we further demonstrate how our method can be used for early forecasting and compare its power against alternative approaches. We also discuss the theoretical implications of the discovered shapes that may help managers identify indicators of a potentially successful movie early and dynamically.