Abstract: This paper explores the relationship between different online advertising and the dynamic transition in consumers’ advertising responses, which are not directly observable, from a hidden dynamic perspective. Using the data of a large e-commerce company in China, it develops a continuous-time hidden Markov model to analyze the behavioral characteristics and dynamic transfer paths of advertising response of consumer groups across different advertising channels. The findings show that most online advertising significantly influence the transition of consumers’ advertising responses. Specifically, customer-initiated contacts advertising has a more pronounced effect on consumer advertising response compared to firm-initiated contacts advertising. Additionally, the non-transactional behavior of consumers under CICs advertising remains basically consistent, while the non-transactional behavior of consumers under firm-initiated contacts advertising differs significantly. From the perspective of dynamic transition, this paper reveals the underlying reasons behind the different effects of various online advertising, and provides guidance for enterprises to effectively place online advertisements.

Key words: FICs online advertising, customer-initiated contacts (CICs) online advertising, continuous-time hidden Markov model, advertising response, dynamic transfer

摘要： 基于隐藏的动态视角，探讨不同在线广告与消费者无法被直接观测的广告响应度变化之间的关系。利用中国一家大型电子商务企业的相关数据，运用连续隐马尔可夫模型分析不同广告渠道下消费者群体的行为特征以及广告响应动态转移路径。研究发现：大多数在线广告均能显著影响消费者广告响应度的动态转移；消费者发起的广告相对于企业发起的广告，对消费者广告响应的提升作用更明显；消费者发起的广告渠道下消费者的非交易行为表现基本一致，而企业发起的广告下消费者的非交易行为有较大的差异。研究从动态转移视角揭示不同在线广告产生不同效果的深层原因，为企业有效投放在线广告提供一定指导。

关键词: FICs在线广告, CICs在线广告, 连续隐马尔可夫模型, 广告响应, 动态转移