非線性電影票房預測模式 : 以美國市場為例

機構典藏 > College of Business and Management > Department of Management Sciences > Thesis > Item 987654321/105325

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/105325

Title:	非線性電影票房預測模式 : 以美國市場為例
Other Titles:	A nonlinear box office forecasting model : evidence from the U.S. market
Authors:	簡妤庭;Chien, Yu-Ting
Contributors:	淡江大學管理科學學系碩士班陳怡妃;Chen, I-Fei
Keywords:	多元適應性雲形迴歸;票房預測;K-means;首週票房;multivariate adaptive regression splines (MARS);box office forecasting;revenue of opening weekend
Date:	2015
Issue Date:	2016-01-22 14:53:16 (UTC+8)
Abstract:	由於消費者需求的變化快速，導致產品的生命週期縮短，新產品的銷售預測成為眾多公司保持競爭優勢的關鍵環節，但如何準確預測銷售一直以來都是困難的問題。電影產業為世界最重要的流行創新產業之一，但其產品生命週期不僅相對其他產業新品短，且每一部新電影的描述資料零星不全，致使預測電影票房的任務更具挑戰性。因此本研究藉由蒐集網路現有口碑資料，及考慮電影的固定效用、首週上映廳數、競爭數、首週票房及季節性等因素，並整合K-Means與多元適應性雲形迴歸（multivariate adaptive regression splines, MARS）以建構電影票房之測預模型，並以美國電影市場為例進行實證分析。結果顯示，本研究所提之無母數適應性預測模型，非但可以解決電影票房異質性高、離散值大的問題，而能獲有高效之預測能力，同時，取得各影響票房重要因素之轉折值，將有利於企業在嚴峻的預算限制下，在短時間內預測新產品銷售或快速進行策略的改善。 Motion picture is one of the most important popular creative industries in the world. Its product life cycle is shorter than other industries and each new film’s data is incomplete which leads to a challenging box office forecast. As the result, we gather the online data for movie MAPP, genres, word of mouth, holiday, the number of opening screens and the revenue of opening weekend as the predictors and combine with K-Means and MARS to construct the box office forecasting model. The empirical results of U.S. market shows that the nonlinear and nonparametric model we used not only can effectively deal with the heterogeneous problem but also can obtain the high-efficiency forecasting performance.
Appears in Collections:	[Department of Management Sciences] Thesis

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	172	View/Open

Loading...