讲座题目：Identification of Advertising Effects

                         
时间：2018年11月21日（周三），12🌰：30  

                     
地点：主教615  

                                     
主讲人💟：厉行，助理教授🧑🏻‍🦽‍➡️，现任教于北京大学光华管理K8凯发营销系♦️🚑，于2016年获得美国斯坦福大学经济学博士学位。厉行的研究使用实证的方法分析不同行业中的经济和营销问题🛁，包括版权保护下的图书定价、快消行业的产品线设计、手机应用的广告策略、组织内部的目标和激励。厉行正在进行的研究包括空气污染下的用户行为👷‍♀️、音乐流应用的用户转换💴、社交网络中的信息传播、以及广告效果的识别。
                     
主要内容：The challenge of measuring advertising effects in the field has spurred new quasi-experimental approaches exploiting the fact that advertising decisions are made at a more aggregate level than which we measure response. We illustrate the common features of these strategies in a simple model of advertising allocation and demand response and then use that model to motivate the application of “Waldfogel” instruments for the measurement of advertising effects. We then assess these and traditional IV approaches on criteria such as i) justifiability of the identifying assumptions within the model, ii) representativeness of the sample used for inference, iii) measurement of marginal vs. inframarginal advertsing effects, iv) how much of actual advertising variation is used for inference, and v) the general applicability of the approach. Using an empirical example, we find that our proposed Waldfogel instruments perform well in most dimensions. Traditional quasi-experimental approaches, such as border approaches, outperform Waldfogel IVs in some dimensions (e.g., iv), but suffer from the dimension of locality (ii), i.e. the estimated effects are “local” to the border. We suggest and illustrate approaches to evaluate local effects in practice.