Welcome! I am a Ph.D. Candidate in Economics at The University of Chicago and work on Economic Theory, in particular Mechanism Design and Game Theory.
I will be participating in the 2019-2020 job market and attending the ASSA 2020 Annual Meeting in San Diego.
Curriculum Vitae: [pdf]
- Credible Auctions with Correlated Values (Job Market Paper) [Draft coming soon]
I consider the auction design problem of maximizing the revenue from selling a good when buyers have correlated values and any mechanism chosen must be credible, in the sense of Akbarpour and Li (2019). Motivated by the rapid expansion in the use of online auctions and the algorithmic black-box in which they operate, credibility requires it be incentive compatible for an auctioneer to follow through with promised rules when bidders take them as given, have private observations in the game, and any deviation not unilaterally detected is feasible. Developing on ideas from the motivating paper, I model the knowledge held by bidders in the run of a promised auction to derive results on dynamic credible implementation of outcomes useful to various design problems with credibility. Turning attention to the correlated values setting, I first show that the optimal credible winner-paying auction when bidders have binary values is found among variations on a first-price Dutch (descending) or second-price English (ascending) auction. Finally, using the implementation results I provide preliminary arguments towards showing the following conjecture: unlike the case of full commitment, there is no mechanism satisfying: (1) bidder incentive compatibility, (2) bidder individual rationality, and (3) credibility that achieves full surplus extraction.
Works in Progress
- Optimal Credible Auctions with Correlated Values
- Mechanism Design for Speech and Debate
- Identification of Average Marginal Effects Under Misspecification when Covariates are Normal
A previously known result in the econometrics literature is that when covariates of an underlying data generating process are jointly normally distributed, estimates from a nonlinear model that is misspecified as linear can be interpreted as average marginal effects. This has been shown for models with exogenous covariates and separability between covariates and errors. In this paper, we extend this identification result to a variety of more general cases, in particular for combinations of separable and non-separable models under both exogeneity and endogeneity. So long as the underlying model belongs to one of these large classes of data generating processes, our results show that nothing else must be known about the true DGP---beyond normality of observable data, a testable assumption---in order for linear estimators to be interpretable as average marginal effects. We use simulation to explore the performance of these estimators using a misspecified linear model and show they perform well when the data are normal but can perform poorly when this is not the case.