Person:
Lai, Zhenyu

Profile Picture

Email Address

AA Acceptance Date

Birth Date

Research Projects

Organizational Units

Job Title

Last Name

Lai

First Name

Zhenyu

Name

Lai, Zhenyu

Filters

2013 - 20152

Settings

Author's Publications: search.filters.isAuthorOfPublication.49a92154-f8f6-4246-902d-edfef640765a

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Publication
    Generalized Random Utility Models with Multiple Types
    (Neural Information Processing Systems Foundation, Inc., 2013) Azari Soufiani, Hossein; Diao, Hansheng; Lai, Zhenyu; Parkes, David
    We propose a model for demand estimation in multi-agent, differentiated product settings and present an estimation algorithm that uses reversible jump MCMC techniques to classify agents' types. Our model extends the popular setup in Berry, Levinsohn and Pakes (1995) to allow for the data-driven classification of agents' types using agent-level data. We focus on applications involving data on agents' ranking over alternatives, and present theoretical conditions that establish the identifiability of the model and uni-modality of the likelihood/posterior. Results on both real and simulated data provide support for the scalability of our approach.
  • No Thumbnail Available
    Publication
    Essays in the Industrial Organization of Internet Markets
    (2015-05-11) Lai, Zhenyu; Pakes, Ariel; Lewis, Gregory; Edelman, Benjamin G.
    This dissertation contains essays that explore the incentives, mechanisms and strategies of platforms in concentrated internet markets. The first essay studies how two-sided platforms compete and set dynamic prices. In the daily deals industry, an expected increase in competition for seller participation prompted a cross-side response by the dominant platform, Groupon, to lower prices on deal coupons sold to attract platform users. I use a dynamic model of competing daily deals platforms to investigate why we observe a larger downward price response in markets where platforms have similar shares of users. Simulations show a sharp increase in value of an additional user as cross-side competition intensifies for profits from being the future platform leader. The second essay---coauthored with Benjamin G. Edelman---examines how competition is mediated by search intermediaries, looking at the design of search engines' own services and the effects on users' choices. We evaluate a natural experiment, and find that Google's prominent placement of its Flight Search service increased clicks on paid advertising listings while decreasing the clicks on organic search listings by a similar quantity. Empirical results and a controlled experiment links the mechanism to users' heterogeneous methods of search. The third essay---coauthored with Michael Egesdal and Che-Lin Su---develops empirical tools for estimating dynamic discrete-choice games. We formulate the maximum-likelihood estimator as a constrained optimization problem to be solved using state-of-the-art constrained optimization solvers. Monte Carlo results show that the constrained optimization approach has improved convergence properties over other popular computational methods.