Skip to content Skip to navigation

RAIN schedule for Spring Quarter 2013-14

The RAIN seminar is held on Wednesdays from 12:00-1:00pm in Y2E2 101. And yes, lunch is provided!

Date Speaker Topic
April 2 Omer Tamuz Strategic Learning and the Topology of Social Networks
April 16 Kostas Bimpikis Information Provision in Dynamic Innovation Tournaments
May 14 Jonathan Levin Sales Mechanisms in Online Markets: What Happened to Internet Auctions?
May 28 Steve Tadelis Quality Externalities and the Limits of Reputation in Two-Sided Markets
June 4 Sinan Aral The Dynamics of Online Reputation

Google Calendar for RAIN

Previous year's talks

Archived talks can be accessed here.

Talk Abstracts

Strategic Learning and the Topology of Social Networks
Omer Tamuz, Microsoft Research/Caltech

We consider a group of people who each have to make a repeated choice between two lifestyle options, and who observe the choices of their friends in a social network. We assume that one of the two options is universally preferable, that each person's choice affects only his or her own health, and that any benefits are only observed in the distant future.

We model this by a Bayesian game of pure informational externalities, in which agents start with conditionally independent private signals. We show that the question of whether or not the agents learn the preferable choice depends on the topology of the social network. In particular, we identify a geometric "egalitarianism" condition on the social network graph that guarantees learning in infinite networks, or learning with high probability in large finite networks, in any equilibrium of the game.

Joint work with Elchanan Mossel and Allan Sly

Bio: Omer Tamuz is a Schramm Postdoctoral Fellow at Microsoft Research New England / MIT, and an Assistant Professor (on leave) at the faculty of the Humanities and Social Sciences division at Caltech. He received his PhD in mathematics from the Weizmann Institute, where he studied learning in Bayesian games, as well as other topics in machine learning, probability theory and ergodic theory.

Information Provision in Dynamic Innovation Tournaments
Kostas Bimpikis, Stanford

Innovation tournaments have emerged as a viable alternative to the standard research and development process. They are particularly suited for settings that feature a high degree of uncertainty about the feasibility of the innovation goal. Information about the progress of participants in these settings has an interesting dual role. On one hand, participants who have experienced no progress start becoming pessimistic about the underlying environment and whether the innovation goal is even feasible, so learning about their competitors' gradual progress provides a positive signal about the attainability of the goal and provides an incentive for participants to continue putting effort. On the other hand, information about the status of competition may adversely affect effort provision from the laggards, who start getting discouraged about their chances of winning the tournament. Because the tournament's success crucially depends on the effort decision of the participants, the tournament's information provision mechanism is a central feature of its design. This paper explores these issues and suggests a number of design guidelines -- with a focus on the role of intermediate awards as information provision devices -- that help implement the objective of maximizing the probability and minimizing the time of completing the end goal.

Bio: Kostas Bimpikis is an Assistant Professor of Operations, Information and Technology at Stanford University's Graduate School of Business. Prior to joining Stanford, he spent a year as a postdoctoral research fellow at the Microsoft Research New England Lab. He received a PhD in Operations Research from the Massachusetts Institute of Technology in 2010, an MS in Computer Science from the University of California, San Diego and a BS degree in Electrical and Computer Engineering from the National Technical University of Athens, Greece.

Sales Mechanisms in Online Markets: What Happened to Internet Auctions?
Jonathan Levin, Stanford

Consumer auctions were very popular in the early days of internet commerce, but today online sellers mostly use posted prices. Compositional shifts in the items being sold and the sellers offering these items cannot account for this evolution. Instead, the returns to sellers using auctions have diminished. We consider two hypotheses: a shift in buyer preferences away from auctions, and an increase in competition or search efficiency that favors favors posted prices. We use a model and data from eBay to tease these apart, finding that the former is more important, especially for more idiosyncratic products. We also discuss why auctions and posted prices can coexist in online markets.

Bio: Jonathan Levin is Professor of Economics at Stanford University. His research interests include industrial organization and microeconomic theory.

Quality Externalities and the Limits of Reputation in Two-Sided Markets
Steve Tadelis, Berkeley

Buyers in two-sided marketplace platforms may draw conclusions about the quality of the platform from any single transaction. This induces an externality across sellers that reputation mechanisms will not alleviate. Furthermore, buyers who abandon the platform without leaving feedback will cause seller reputations to be biased. Using data from eBay, we document this externality and argue that platforms can mitigate it by actively screening sellers and promoting the prominence of better quality sellers. Exploiting the bias in feedback, we create a measure of seller quality and demonstrate the benefits of our approach through a controlled experiment that prioritizes better quality sellers to a random subset of buyers. We thus highlight the importance of externalities in two-sided markets and chart an agenda that aims to create more realistic models of two-sided markets.

Bio: Steve Tadelis is a professor of economics at Berkeley's Haas School of Business. From 2011 to 2013 he spent a two year leave at ebay Labs, where he put together and led a group of Economists who focus on the economics of e-commerce, with particular attention to creating better matches of buyers and sellers, reducing market frictions by increasing trust and safety in eBay's marketplace, understanding the underlying value of different advertising and marketing strategies, and exploring the market benefits of different pricing structures. Aside from the economics of e-commerce, Steve’s main fields of interest are the economics of incentives, industrial organization and microeconomics. His research has focused on a firm's reputation as a valuable, tradable asset; the effects of contract design and organizational form on firm behavior with applications to outsourcing and privatization; public and private sector procurement and award mechanisms; and the determinants of trust.

The Dynamics of Online Reputation
Sinan Aral, MIT

Identity and reputation drive some of the most important decisions we make online: Who to follow or link to, whose information to trust, whose opinion to rely on when choosing a product or service, whose content to consume and share. Yet, we know very little about the dynamics of online reputation and how it affects our decision making. Sinan will describe a series of randomized experiments that explore the population level behavioral dynamics catalyzed by identity and reputation online. He will explore some of the implications for bias in online ratings, the foundations of social advertising and the ability to generate cascades of behavior through peer to peer social influence. The coming decades will likely see an emphasis on verified identities online. Sinan will argue that a new science of online identity could help guide our business, platform design and social policy decisions in light of the rising importance of online reputation.

Bio: Sinan Aral is a Scientist and an Entrepreneur: He is the David Austin Professor of Management at MIT, where he holds joint Associate Professorships in the IT and Marketing groups and co-leads the Initiative on the Digital Economy. He was the Chief Scientist at SocialAmp, one of the earliest social commerce analytics companies (until its sale in 2012); and is currently the Chief Scientist at Humin, a social navigation startup developing the “Google Maps” for your social relationships.

Sinan is the Scholar-in-Residence at the New York Times R&D Lab and has worked closely with Facebook, Yahoo, Microsoft, Nike, IBM, Intel, Cisco, Oracle, SAP and many other leading Fortune 500 firms on realizing business value from social media and IT investments. His research has won numerous awards including the Microsoft Faculty Fellowship, the PopTech Science Fellowship, an NSF CAREER Award and a Fulbright Scholarship. He was also recently named one of the “World’s Top 40 Business School Professors Under 40” by Poets & Quants.

In his spare time, he cooks, skis and tell jokes about his own cooking and skiing. His most recent hobby is learning from his one year old son.

You can find Sinan on Twitter @sinanaral.