Determination of Investment Success and its Factors for Russian Cinema at the Box Office Using Machine Learning

Abstract

Historical data of the box office of Russian cinema is the object of research. The purpose of the study is to determine the possibility of forecasting the cash fees of the film project at an early stage in the production of films, which is especially important due to withdrawal of foreign distributors from the Russian market. The analysis was carried out on data for the entire population (N = 1400) of Russian national films that were released from the beginning of 2004 to April 2022. These data are introduced into scientific circulation for the first time. The study used methods of evaluation of film projects based on historical profitability and classification of films by genres, directors, screenwriters. The result of the experiment on 7 machine learning and neural network models achieved an accuracy of 0.96 and ROC (AUC) = 0.98. The article provides conclusions about the directions for improving forecasting methods and conclusions about the limitations of the proposed approach. Taking into account the high volatility of the individual financial result of a film project, recommendations were made by the “portfolio” principle of investment, which opens the prospects of debt and equity financing of cinema using market financial instruments, issuance of bonds and shares by producers and distributors.