讲座预告|姜正瑞：Time to Upgrade under Successive Product Generations: A Survival Model with Exponential-Decay Baseline Function 多代产品选择下用户产品升级决策研究——基于指数衰减的生存分析

主题：Time to Upgrade under Successive Product Generations: A Survival Model with Exponential-Decay Baseline Function 多代产品选择下用户产品升级决策研究——基于指数衰减的生存分析

主讲人：姜正瑞

时间：2018年5月9日 上午10:00-12:00

地点：25教学楼C区412

主讲人介绍：

姜正瑞博士是爱荷华州立大学商学院信息系统与商务分析系教授。姜博士的研究领域包括：Business intelligence/analytics, Data quality, Decision-making under uncertainty, Diffusion of technological innovations,以及Economics of information technology。他的多项研究成果已经发表在顶级学术期刊上,包括Management Science, MIS Quarterly, Information Systems Research, INFORMS Journal on Computing, IEEE Transactions on Knowledge and Data Engineering,和Journal of Management Information Systems等，以及顶级的学术会议上如ICIS, WITS,和INFORMS等。

姜博士目前担任Information Systems Research的副主编。在过去的几年里，他曾担任MIS Quarterly的副主编，以及POM和ACM Transactions for MIS的特刊编辑。 他于2016年获得MISQ杰出副编辑奖。此外他还担任第十三届年度Big XII+ MIS研究研讨会（2015）和第九届中西部信息系统年会（2014）的联合主席。他目前担任第28届信息系统与技术研讨会（WITS）的联合主席。姜博士的研究已经被美国国际开发署（USAID）和中国国家自然科学基金会（NSFC）资助。

讲座介绍：

In the presence of successive product generations, most customers are repeat buyers, who may decide to purchase a future product generation before its release. As a result, after the new product generation enters the market, its sales often show a declining pattern, making traditional bell-shaped diffusion models unsuitable for characterizing the timing of product upgrades by customers. In this study, we propose asurvival model with exponential-decay baseline function(orexponential-decaymodel) to predict customers’ time to upgrade to a new product generation. Compared with existing proportional hazard models, the exponential-decay model is parsimonious and easy to interpret. In addition, empirical analysis using upgrade and usage data for a major sports video game series shows that the exponential-decay model performs better than or as well as existing parametric models in prediction accuracy. Furthermore, we show that, by extending the basic exponential-decay model to a time-variant model, the prediction accuracy can be further improved. Empirical results obtained using the basic and time-variant exponential-decay models are quite consistent, both revealing that customers’ previous adoption and usage patterns can help predict their timing to upgrade to a new product generation. In particular, we find that (i) potential customers who have adopted the previous generation are more likely to upgrade; (ii) heavy users tend to upgrade earlier; (iii) specialized customers demonstrate a lower probability to upgrade. These findings can help firms better understand customers’ upgrade behaviors and develop more personalized promotions to target customers.