题  目：How Mega is the Mega?

Measuring the Spillover Effects of WeChatUsing Graphic Model

报告人：窦一凡副教授   复旦大学管理学院

时  间：2021.12.1 10:30-11:30

地  点：线上腾讯会议419 440 452

Abstract

WeChat, an instant messaging app, isconsidered a mega app because of its dominance in terms of use among Chinesesmartphone users. Little is known, however, about its externality in thebroader app market. This work estimates the spillover effects of WeChat on theother top 50 most frequently used apps in China, using users’ weekly app usagedata. Given the challenge of determining causal inference from observationaldata, we apply a graphical model and an econometric method to estimate thespillover effects in two steps: (1) we determine the causal structure byestimating a partially ancestral dia- gram, using a fast causal inferencealgorithm; and (2) given the causal structure, we find a valid adjustment setand estimate the causal effects by an econometric model with the adjustment setfor controlling noncausal effects. Our findings show that the spillover effectsof WeChat are limited; in fact, only two other apps, Tencent News and Taobao,receive positive spillover effects from WeChat. In addition, we show that ifresearchers fail to account for the causal structure that is determined fromthe graphical model, it is easy to fall into the trap of confounding bias andselection bias when estimating causal effects. The findings generate managerialimplications in terms of app usage patterns, strategic management of mega appson an app platform, and app promotional strategies for app platform managersand app developers.

主讲人介绍：窦一凡，复旦大学管理学院信息管理与信息系统系副教授、博士生导师。他的主要研究兴趣是电子商务、平台模式和数字虚拟产品的经济学问题，他的研究受到国家自然科学基金优青项目资助。他目前是信息系统领域国际顶级期刊《Information Systems Research》的Associate Editor。