2020年11月26日
2020年11月26日
最新消息:
您现在的位置:首页 >> 论坛讲座

上海市市场学会参与主办的“华人学者营销协会第六届中国市场营销国际学术年会”分会场 · 第五辑

第五辑 | 大数据营销分会

华人学者营销协会第六届中国市场营销国际学术年会(China Marketing International Conference 2018?)将于2018年7月20-22日在上海财经大学举行

会议以“智慧营销:人、技术与创新”为主题,由华人学者营销协会、上海财经大学商学院、上海市市场学会共同主办

CMIC设有9个特邀分论坛,分别代表营销研究和实践领域的9个专题。大会邀请与主题相关的优秀青年学者和业界人士进行专题分享和交流。这些分会场时间安排在7月21日和7月22日下午,欢迎您选择自己感兴趣的专题前往。



今天,

小编为大家介绍第五辑内容。

第五辑 大数据营销分会

?

7月22日下午13:00-17:00?

?

大会现场通知

分会主席


董晓静教授(圣塔克拉拉大学)
??


演讲人介绍


罗学明

罗学明是天普大学福克斯商学院Charles Gilliland营销学杰出主席教授、策略和信息管理教授、移动分析大数据全球中心的创始人。


罗教授的研究方向主要是利用机器学习、大数据和大规模田野实验研究移动营销、消费者分析、数字创新、共享经济平台以及社会责任策略。他专注于研究如何将工程模型和田野实验方法应用于消费者洞察、市场研究和商业策略。罗教授目前的研究领域包括:移动端消费者分析、广告精准投放的深度学习、竞争定价、全渠道营销、社交媒体网络广告、人工智能和平台算法以及顾客绩效的财务价值。


罗学明教授合作过的企业包括全球领先的移动通信公司、银行、零售公司、健康医疗、医药公司以及石油产业的公司。他的研究不仅发表在营销、策略、信息系统以及管理领域的顶级期刊上,也发表在Wall Street Journal、ScienceDaily、Forbes、Financial Times、Harvard Business Review、MIT Sloan Management Review等著名商业媒体刊物上。


胡曼恬

胡曼恬现任香港中文大学工商管理学院市场学系助理教授。本科毕业于复旦大学,于纽约大学Stern商学院取得博士学位。


胡教授的主要研究方向是运用前沿的实证研究方法进行数据分析和挖掘从而探索和解释在TMT等行业中的消费者行为,特别是社交网络,口碑效应以及人际互动在营销活动中的作用及影响。其研究成果发表于Marketing Science、The International Journal of Research in Marketing等国际顶尖营销类学术期刊。她曾获美国“Society for Marketing Advance 学会博士论文竞赛最佳论文”奖。


胡教授曾担任香港数码分析协会荣誉顾问并为国内外市场研究公司,电信企业以及手机制造商提供营销策略咨询。


李洋

李洋博士现任长江商学院副教授。他本科毕业于北京大学电子学系,之后取得美国哥伦比亚大学生物医学工程硕士,哥伦比亚大学商学院博士。李洋专注于营销大数据模型开发和公司大数据战略实现。李洋博士在营销数据模型、人工智能算法等方面的研究成果已发表在Management Science, Marketing Science, Journal of Marketing Research等管理类国际A级学术期刊上,常在美国和欧洲的学术机构做关于数据模型的演讲。在长江商学院李洋讲授EMBA、EE、FMBA和MBA等项目课程,曾为腾讯、百度、永辉超市、海尔、海底捞等企业提供品牌战略咨询,并持有医学图像处理的美国专利。


?沈俏蔚

沈俏蔚现任北京大学光华管理学院营销学教授。她于加州大学伯克利分校取得营销学博士学位,于北京大学取得经济学学士和硕士学位。


沈教授的研究方向包括实证企业建模、消费者决策、社交互动和新媒体以及竞争营销策略。她的研究成果发表在Marketing Science和Management?Science等顶级期刊上,曾获得过国家自然科学基金颁发的2017年优秀青年学者的荣誉。


在加入光华管理学院的教师队伍之前,沈教授还曾在宾夕法尼亚大学沃顿商学院任教。


董晓静

董晓静现任圣塔克拉拉大学利威商学院营销学与商业分析副教授,也是利威商学院商业分析硕士的创始人,并担任学位总监。她于美国西北大学取得博士学位,于MIT取得硕士学位,于清华大学取得学士学位。


董教授的研究方向包括消费者精准分析、消费者决策过程研究以及商业行为对于消费者决策的影响等。她的研究成果发表于Marketing Science、Journal of Marketing Research、Quantitative Marketing and Economics等顶级期刊上。她曾于2005年获得美国营销科学学会颁发的Alden Clayton奖。


分享内容摘要


罗学明教授分享内容摘要如下:

Popularity-based or Personalization-based Algorithms for the Sharing Economy Platform?

?Evidence from Natural Experimentation and Machine Learning

?

Xueming Luo (Temple Univ),?

Zhijie Lin (Nanjing Univ),?

Siliang Tong (Temple Univ),

?Jing Li (Hong Kong Polytechnic Univ)

?

This paper examines how platform recommendation algorithms based on demand-side tastes affect supply-side small scale entrepreneurs. It addresses three vital questions: (1) What is the effect of implementations of popularity or personalization platform recommendation algorithms on seller revenue? (2) How do changes of platform algorithms incentivize sellers in the sharing economy? (3) For which sellers the platform algorithms are more beneficial? Through natural quasi-experiments and rich proprietary datasets from a major food-sharing platform, the analysis finds significant increases of seller revenues after the platform implements either algorithm. But, the pathways to these revenue increases differ. As the review popularity recommendation (RPR) platform algorithm helps buyers to find sellers with high review ratings more easily, sellers are incentivized to adopt a specialization?focus on the quality reputation of current products. By contrast, as the bolter personalization recommendation (BPR) algorithm enables buyers to find sellers with more customized cuisines, sellers respond by adopting an innovation?focus on introducing more new products to suit the diverse customer tastes. Consistent with the specialization pathway, RPR is more beneficial for sellers who have a concentrated product assortment. In contrast and in line with the innovation pathway, it is younger and newer entrepreneurs that reap more benefits from BPR. Surprisingly, each algorithm has unintended outcomes: RPR impedes innovation and BPR inhibits specialization. However, the platform manager can leverage a machine learning causal forest technique to learn sellers’ heterogeneous responses to RPR and BPR and craft an optimal targeting rule, which maximizes algorithms’ benefits and minimizes their negative effects for the sharing platform.


胡曼恬教授分享内容摘要如下:

Recommendation Systems for Sequential Decisions with Limited Time Offers

?

Mantian HU?(The Chinese University of Hong Kong),?

Tong Zhao?(Amazon.com, Inc.),?

Irwin King?(The Chinese University of Hong Kong)

?

Although literature on recommendation systems in marketing has demonstrated the value of introducing consumer behavior models to improve the performance, most of them have focused on the situation in which choices are simultaneously available such as recommendations on Amazon. In this paper, we propose a recommendation system for sequential decision-making scenarios in which choices are time sensitive (will expire in a short time) and made sequentially available and the quality of future choices is uncertain, for example, recommendations on daily deal websites. The proposed recommendation system is grounded in the search literature of Economics and we develop a new machine learning algorithm to solve the search model. In particular, we leverage the means of Semi-parametric Gaussian copula (SGC) to generate a complex joint distribution among variables including individual characteristics, choice features, sequential temporal factor and purchase decisions. Recommendations are made based on the purchase likelihood of each available deal on a specific day for an individual calculated by the conditional posterior of the joint distribution. The authors apply the model to a proprietary dataset from Groupon customers' clickstream data and demonstrate the superior performance of the proposed approach by comparing it to other traditional or sequential based recommendation algorithms, including collaborative filtering methods and multi-armed bandit models. The system facilitates companies' efforts to target the right customers with the right product at the right time.


李洋教授分享内容摘要如下:

Dynamic Preference Heterogeneity

?

Asim Ansari (Columbia University),

Yang Li (CKGSB),

?Ryan Dew (Wharton, University of Pennsylvania)

?

Consumers' preferences and sensitivities to marketing variables change over time, often in tandem with population trends, but frequently exhibiting individual-specific idiosyncrasies. While much of the empirical marketing literature has focused on capturing cross-sectional heterogeneity, little research has been done on modeling the temporal evolution of heterogeneity. In this work, we develop a Bayesian non-parametric framework based on Hierarchical Gaussian Processes (HGP) for modeling dynamic heterogeneity, which flexibly captures both the evolution of population trends and individual-level departures from those trends over time. This novel specification allows for sharing of statistical information across individuals, and within individuals over time, to provide rich individual-level insights and efficient inferences regarding dynamics. We showcase our HGP specification in a choice modeling context, using both simulations and an application to consumer packaged goods data. We find that restricted heterogeneity specifications, as have been employed previously in the literature, can lead to significant biases in the presence of dynamic heterogeneity, even in estimating population-level trends. Moreover, these restricted specifications cannot capture managerially-relevant patterns of individual-level variation around population trends. In our application, we show robust evidence of dynamic heterogeneity across CPG categories during the Great Recession, and illustrate the clear gains from capturing dynamic heterogeneity through our HGP specification. We uncover important individual-specific trends that can be used for targeting, including variability in consumer responses to the recession, and show that targeted pricing that leverages dynamic heterogeneity can lead to higher retailer profits.


沈俏蔚教授分享内容摘要如下:

Whose Opinion is More Valuable? Those who Bought a Lot or Those who Rated a Lot?

?

Li Wang (Shanghai University of Finance and Economics),

?Qiaowei Shen (Guanghua School of Management, Peking University),?

Yuxin Chen (New York University Shanghai)

?

Many firms gather opinions from online users which include not only customers who actually purchase but also individuals who simply voice. Should firms only listen to the buyers? In this paper we address this question by examining a scenario in which a firm uses online votes to determine what products to launch. By analyzing a data set that tracks both individual ratings to product designs and their purchase behavior as well as the eventual market performance of the launched products, we find that both rating experience and purchase experience of the voter group enhance average rating's predictive power of sales. Further analysis reveals that rating experience could be an effective way of learning in terms of discriminating quality of product designs. At the individual level, we find that rating experience makes one's score more detached from his own preference while purchase experience does the opposite. Such difference also explains the finding that purchase experience enhances prediction of sales from core users while rating experience improves predictions of sales from general buyers.


董晓静教授分享内容摘要如下:

Estimation of Preference Heterogeneity in Markets with Costly Search

?

Xiaojing Dong (Santa Clara University),?

Ilya Morozov (Stanford University),?

Stephan Seiler (Stanford University) ,

Liwen Hou (Shanghai Jiaotong University)


We study the estimation of preference heterogeneity in markets where consumers engage in costly search to learn product characteristics. Costly search amplifies the way in which consumer preferences translate into purchase probabilities, generating a seemingly large degree of preference heterogeneity. We develop a search model that allows for flexible heterogeneity in preferences and estimate its parameters using a unique panel dataset on the search and purchase behavior of consumers. The estimation results reveal that ignoring search costs leads to an overestimation of standard deviations of product intercepts by 30 percent. We show that this bias leads to incorrect inference about price elasticities and mark-ups of sellers and has important consequences for optimal targeted marketing.


相关阅读:

会分场预告 · 第四辑?| 数字营销和新媒体分会

分会场预告 · 第三辑 | 共享经济

分会场预告 · 第二辑 | 产品管理与创新
分会场预告 · 第一辑 | 数字化内容营销
会议通知 | 第六届中国市场营销国际学术年会


更多论坛分会场的信息将陆续推出 ??请关注微信公众号