Forecasting movie demand using exponential smoothing and Delphi methods

This study investigated the quantitative and qualitative methods in forecasting movie demand which are total and split exponential smoothing and Delphi methods respectively. In quantitative study, the performance of total and split exponential smoothing was evaluated and compared to other exponential smoothing methods in movie demand forecasting at the aggregate market level. Exponential smoothing methods were suggested because of their simplicity in formulation and generation of reliable forecasts. The forecasts can be generated with minimal effort in the formulation, thus, it tends to shorten the time to make a decision. Nevertheless, there is a limited application of exponential smoothing methods in movie demand forecasting. The model fitting criterion was also examined to see which criteria generate better forecasts. The data were daily sales series of movie market started from 1 January 2002 to 31 December 2016. Overall, the total and split exponential smoothing with optimised parameters was the best performing model. The identification of the best performing method assists distributors to make a decision on the best release date for their new movies earlier than the competitors. The forecasts generated able to give a general picture regarding the future trend of movie demand at the aggregate market level. In a separate study, the qualitative study used the Delphi method to estimate the movie demand at individual movie level. Past research in motion picture claimed the great uncertainty of individual movie demand because of limited information. So, they suggested relying on judgements and intuition as inputs in the forecasting process when there is minimal data condition. Until now, there is no study using the judgemental method in demand forecasting at individual movie level. Eleven movies released in 2017 were selected. Results suggested that, at the individual movie level, the group produced better forecasts than an individual member with the Delphi method. There is an improvement in forecasting accuracy over the Delphi rounds. Lastly, under the condition of great uncertainty, the combined forecasts generated better accuracy over the individual methods. With the proven benefits of Delphi method under great uncertainty of individual movie demand, it was able to give confidence for distributors and exhibitors to rely on judgemental methods other than statistical methods alone