口碑效果之異質性分析-以美國電影為例

Abstract

創新產品的銷售對於成功的企業來說，佔了總銷售額超過60％，因此，新產品的發展為企業是否生存極大關鍵。在過去，受限於多種無法量化的外部環境因素，預測產品生命週期與銷售量的科學模型難以設立，但隨著Bass擴散模型的發展，預測新產品生命週期與銷售量日趨穩定，但以往Bass 擴散模型多以耐久消費財作為研究對象，且又根據美國電影協會(MPAA, Motion Picture Association of America) 發布的資料顯示，電影產業在全球經濟中扮演著高度的重要性，電影票房的總收入逐年升高，近幾年美國票房帄均收入約為90億美元，全球票房總收入約為110億美元，可見電影產業具有龐大的市場潛能，也是美國一大重要市場，因此本論文將採用在美國上映之電影做為實證分析的對象。 近幾年開始，美國學術界開始對電影產業有相關的探討，但多以評價的質與量來衡量其口碑效果，本研究利用觀眾首週對於該部電影的評價之機率分配，詴圖以新的方法來衡量口碑效果。本研究分析模式的主要分析方法為Bass擴散模式，在Bass 擴散模式中加入觀眾對於該部電影評價的機率分配變數，比較舊有只考慮評價質與量的擴散模式，以求最佳創新產品銷售預測模式。

Sales of innovative products for the successful enterprises accounts for more than 60% of total sales, therefore, new product development is the key to success for a company in the modern business world. Being limited by various non-quantifiable external environmental factors during the past, it was not easy to set up the science relation for sales amount prediction. But following Bass model’s development, the model which used to forecast life cycle of new products and sales is more stable. Most former research paper used durable goods as research object, but according to Motion Picture Association of America (MPAA, Motion Picture Association of America), the film industry in the global economy plays an important role, the film's total box office revenue about 11 billion U.S. dollars, and the U.S. box office in recent years about 9 billion U.S. dollars. The film industry is a high potential market, therefore, it used the films which released in U.S. be research object. In recent years, there are more research related to film industry, but most of them used quality and quantity to measure word-of-mouth effect. This study added “the probability distribution of the evaluation of audience for the film in the first week” factor to measure word-of-mouth. The main analytical methods for the analysis model of this research are Bass Diffusion Model. In this research, the probability distribution of the evaluation of audience is added in the Bass Diffusion, as such, it is the purpose of this research to obtain optimal innovative product sales amount prediction model for providing a model example of scientific prediction to the industry.