Research
Working Papers
Innovations, either products or ideas, often diffuse in the population via social ties. This paper studies the identification and estimation of diffusion processes in social and economic networks. Compared to the classic econometric diffusion literature that assumes a continuous population with a stochastic network structure, we provide a new econometric framework to analyze diffusion processes in fixed networks where Bayesian players observe their close neighbors. We demonstrate the existence of the equilibrium of the model and characterize the symmetric equilibrium. Based on these theoretical findings, we propose a consistent and tractable two-step estimator for payoff parameters using feasible data from a single large network. We evaluate the finite sample performance using Monte Carlo simulations, and apply our method to the network data from Banerjee et al. (2013).
"The Identification in Many-to-Many Matching Games" (Draft)
This paper analyzes identification in a random utility model of many-to-many matching markets, where agents on both sides can form multiple links and market sizes grow at potentially different rates. The number of links formed by each agent is endogenously determined within the equilibrium.
I propose a utility structure that satisfies the stability conditions necessary for many-to-many matchings and is tractable for empirical implementation. A central challenge in these markets is that expanding the capacity of agents on either side can significantly affect the set of stable outcomes. I show that, under specific conditions, the number of firms willing to match with a given worker (and vice versa) grows at a rate independent of capacity expansions. This property enables the use of the inclusive value framework introduced by Menzel (2015) to characterize agents’ opportunity sets and facilitate identification.
"Wholesale Spot Markets and Retail Energy Supply" with Kelly Neill
For commodities where supply is dominated by long-term contracts, introducing a wholesale spot market can reduce overall supply to final consumers. We propose a model of retail and distribution firms with market power, that acquire a commodity in sequential markets: long-term contracts and spot, and then sell it to final consumers. We show that adding a spot market can help firms commit to lower overall supply quantities.
Uniquely, Australia has introduced daily spot markets for natural gas flows at physical hubs, with rules similar to wholesale electricity markets. Recently, fringe firms have increased supply in these spot markets. At the same time, large retail firms have reduced supply to industrial gas users. We apply our framework to investigate whether the two are linked. To do this, we use spot market data to estimate the firm-level supply to final users, as well as costs under long-term contracts with upstream producers.
Work in Progress
"Bank Runs with Social Networks" with Danxia Xie, Zhihao Xu
The study examines the phenomenon of bank runs in the context of social networks. It investigates how social connections influence the likelihood of bank runs, focusing on their impact on individual behavior.