Possible Forecasting Method for Box Office

DOI

10.2991/icesed-19.2020.17How to use a DOI?

Keywords

Box office, Multiple linear regression, Random forest, Recurrent neural network.

Abstract

The booming development of film industry and the emergence of high-return films have attracted the attention of investors. It is well known that high returns mean high risks. In order to avoid risks, scholars and practitioners have studied various box office prediction models. This paper elaborates the thought about the possible forecasting method for box office. At first, the author selects some features based on the experience and previous research. Next, some traditional features are removed and redefined. After determining the original influencing factors, multiple linear regression and random forest selection are used to select the factors with strong significance. Finally, the author chooses films which are not in the sample range and takes their data into recurrent neural network to get the predicted results. Getting the consequence, the author compares with the actual box office to confirm the effectiveness of the method. This paper is expected to provide readers with a research idea.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xinyi Zhang
PY  - 2020/01
DA  - 2020/01
TI  - Possible Forecasting Method for Box Office
BT  - Proceedings of the 2019 International Conference on Education Science and Economic Development (ICESED 2019)
PB  - Atlantis Press
SP  - 292
EP  - 298
SN  - 2352-5428
UR  - https://doi.org/10.2991/icesed-19.2020.17
DO  - 10.2991/icesed-19.2020.17
ID  - Zhang2020/01
ER  -