A generalized Bass model for predicting the sales patterns of motion pictures having seasonality and herd behavior

Abstract

Marketing scholars have expended much effort in developing quantitative models that forecast box office results, but performance has been below their expectations because of the failure to notice behavioral factors, such as herd behavior. Prior studies also have relied on weekly box office data, not daily data, and thus have failed to provide marketers with sufficient insights necessary for making key decisions. This paper offers a box office forecasting model that overcomes these weaknesses: a generalized Bass model (GBM) that reflects both daily seasonality and herd behavior. The daily box office data of 40 movies released in Korea are used for the analysis. In all, three models are compared: a GBM that takes into consideration both herd behavior and seasonality, a GBM considering seasonality only, and an exponential model as a benchmark model. The estimation results show that the proposed model performs better than the others in terms of model fitness and predictability. Keywords: demand forecast; box office; movie marketing; seasonality; herd behavior 市场营销学者在发展定量模型预测票房结果上做了很大的努力, 但是表现低于他们的预期, 因为 没有注意到从众的行为因素。之前的研究也是依靠周票房数据而不是日票房数据, 因此对于制 定关键的市场决策不能提供足够的洞察力。 本文的目的是提供一个克服先前局限性的票房预测模型--通用的 Bass 模型反映了日常 的季节性以及从众行为。此模型利用电影发行首日的票房数据。当然, 这将是在电影上映前很 理想的预测准确需求。然而, 为了更好地预测票房结果, 利用早期有效的数据是很有必要的。考 虑到电影平均放映期会超过六周, 如果在电影放映数周后可以预测较为准确的销售模式是有益 的。在目前的研究中, 使用了40 部韩国电影的日票房数据。三个模型进行了比较：GBM(通 用的 Bass 模型)同时考虑到从众行为和季节性, GBM是一个指数模型只考虑季节性。电影发 行后14天内的销售数据被用作模型的校准, 预测35天的票房销售情况。预测结果显示, 在适度 分析和预测性方面, 此模型的表现要好于其他模型。 所有三个模型显示了较好的模型拟合。在 Adjusted-R2 项中, GBM-SH 是最高值为 0.963, 紧随其后的是GBM-S (0.958), 其次是 EXP-S (0.907)。为了比较模型的适合度, 经行类似然比 检测。GBM-S比 EXP-S 更为显著, GBM-SH 要强于 GBM-S。通过衡量平均绝对误差百分比的 ISSN 2163-9159 print/ISSN 2163-9167 online q 2012 Korean Scholars of Marketing Science http://dx.

The diffusion pattern of new products: evidence from the Korean movie industry

“…This origination of the Bass model indicates that the model is suitable for describing a diffusion pattern of a product category having a long-life span, such as refrigerators, televisions, air conditioners, and irons. However, the model has also been successfully applied to movie-going behavior because of its relevance to the movie selection process and resemblance to single-purchase characteristic (Chance et al, 2008 ; Dellarocas et al, 2007 ; Lee et al, 2012 , 2017 , 2019 ; Marshall et al, 2013 ; Sawhney & Eliashberg, 1996 ). Even though a single movie is not a product category but a single product, a movie has some common features that consumers can be grouped by when they adopt a new product, and the non-adopters are affected by the previous adopters through word-of-mouth.…”

“…If new product diffusion pattern changes over time in a way that shows a growing level of the Bass model coefficients, diffusion of movies would not be an exception. However, even though various studies have applied box office data to the Bass model (Dellarocas et al, 2007 ; Lee et al, 2012 , 2017 , 2019 ; Marshall et al, 2013 ; Sawhney & Eliashberg, 1996 ; Zhang et al, 2020 ), studies focusing on chronological changes of movies diffusion are extremely scarce. Therefore, a meta-analytic approach is required to identify the evidence of the historical change in a single market.…”

Big Social Data Analytics in Football: Predicting Spectators and TV Ratings from Facebook Data

“…Researchers have already utilized big social data to predict stock market movements (e.g. [12], [13]), announcements of flu outbreaks [14], forecast revenues for movies ( [15], [16]) and to predict election outcomes [17]. Lee, Kim and Cha [16] used a generalized Bass Model (GBM) that reflected both daily seasonality and herd behavior to predict the sales patterns of motion pictures.…”

“…[12], [13]), announcements of flu outbreaks [14], forecast revenues for movies ( [15], [16]) and to predict election outcomes [17]. Lee, Kim and Cha [16] used a generalized Bass Model (GBM) that reflected both daily seasonality and herd behavior to predict the sales patterns of motion pictures. This is also an interesting model for this paper since football matches might also experience daily seasonality (with higher attendance at matches played on weekends and holidays) and herd behavior.…”

Identifying the Diffusion Patterns of Movies by Opening Strength and Profitability

“…확산 패 턴에 관한 연구는 상품이나 서비스의 특성을 이해하고 그에 따 른 정책 도입 등 전략적 의사 결정의 근거가 된다는 점에 중요 한 의의를 가지고 있다. 이 주제에 관한 연구는 Teng et al (2002), Kim and Kim(2006) 등에서 특히 혁신적인 IT 기술의 확 산 과정을 중심으로 다양하게 이루어지고 있으며, 유사한 형태 의 연구가 같은 문화 산업의 특성을 공유하는 책이나 음반에 관 해서 이루어진 바가 있으나 (Beck, 2007) (Park, 2002;Dellarocas et al, 2007;Chang, 2008;Lee, 2011;Lee et al, 2012 (Ravid, 1999;Dellarocas et al, 2007;Qin 2011 (Hong and Eom, 2009;Koo et al, 2011). 외부 효과와 내부 효과는 각 각 혁신과 모방으로도 불리며, 다음의 미분 방정식에서 두 계 수 p, q로 표현된다.…”