An Empirical Study Of Movie Box Office Prediction

In recent years,the film industry in our country has been developing rapidly.The box office income and the number of people watching are also increasing year by year.National policies have repeatedly identified the important position of the cultural industry in the development of the national economy.As an important part of the cultural industry,the film industry's development status is self-evident to China's economic development.As a high-risk experiential product,investors often need to consider risks and benefits to decide whether to invest in the film.In general,investors mainly consider it as the main indicator of return on investment-box office income.If you can make reasonable and accurate predictions at the movie box office,you can avoid investment risks to a great extent,and you can make reasonable adjustments to the strategy in the filming and production of the film,as well as the propaganda and screening process in the later stage,and maximize the investment income..The reasonable and accurate prediction of movie box office not only can promote the development of China's cultural undertakings,but also promote the development of China's national economy.On the basis of sorting out the relevant literature,this paper constructs a movie box-office forecasting index system consisting of 17 variables including publishers,series of works and 13 categorical variables,as well as the director's box office appeal and number of announcements.A total of 305 videos released in mainland China between August 2016 and August 2017 were selected for prediction analysis.Using SPSS software,combined with multiple linear regression analysis and neural network analysis to fit the sample data prediction.Prior to multiple regression,variables were initially screened using one-way ANOVA and linear regression with one variable,and a total of 14 variables were selected for multiple linear regression fitting.Finally,according to the results of the empirical analysis,several feasible suggestions were made for movie publishers and movie theaters.The results show that the main box office,series of works,directors and starring box office appeals,publicity articles and viewers want to see the index will have a significant positive impact on the final box office movie revenue;science fiction films will have a significant negative impact on the final box office income.Among the many factors affecting movie box office,according to the importance of variables in the model,the order is the director's box office appeal,starring box office appeal,number of announcements,general holidays,and import purchase.