Online Program Home
My Program

Abstract Details

Activity Number: 602 - Game Analytics: How Data Science Transforms the Game Industry
Type: Topic Contributed
Date/Time: Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Learning and Data Science
Abstract #304993
Title: Product Diffusion on a Dynamic Matching Platform: The Case of a MMOG
Author(s): Chenyu Yang*
Companies: University of Rochester
Keywords: dynamic matching; structural model
Abstract:

This paper studies consumer learning on a dynamic matching platform. Our empirical context is a massively multiplayer online game with 80 million daily active users and 200 million registered users, where the platform matches active users in battles. We first provide evidence of product experimentation and quasi-experimental evidence of observational learning. We next examine whether learning generates efficient product diffusion. Consumers learn about the private value of a product through active experimentation and observational learning. Using a product with a high private value in a match also increases the match value to the matched partner. Private experimentation does not internalize the benefit of a high quality match on the matched partner and may result in under-learning, an outcome partly mitigated by observational learning. We estimate a structural model of the consumer learning process. In the counterfactual, we examine to what extent reducing prices and seeding can improve the learning outcomes.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program