Ph.D. Candidate in Economics at Cornell University


I am an econometrician broadly interested in the econometrics of networks and structural modeling. I am particularly interested in answering economic questions using models that make realistic assumptions about the observed data. Typically data are limited by the way that they are collected and often contains interesting patterns that standard economic models cannot rationalize. For example, in my job market paper I address the very salient problem where: (1) social networks are not fully observed as a result of subsampling, which has implications for identification of network statistics, and (2) social networks are formed endogenously. As another example, in a recent working paper with Levon Barseghyan and Francesca Molinari that is under revise and resubmit at American Economic Review, we propose an economic model of discrete choice that can explain prominent data patterns that are present in virtually all insurance datasets. These data patterns cannot be rationalized by either the standard expected utility framework or alternate behavioral models proposed in the literature.

Job Market Paper

Identification and Estimation of Network Statistics with Missing Link Data.

I obtain informative bounds on network statistics in a partially observed network whose formation I explicitly model.  Partially observed networks are commonplace due to, for example, partial sampling or incomplete responses in surveys. Network statistics (e.g., centrality measures) are not point identified when the network is partially observed.  Worst-case bounds on network statistics can be obtained by letting all missing links take values zero and one. I dramatically improve on the worst-case bounds by specifying a structural model for network formation.  An important feature of the model is that I allow for positive externalities in the network-formation process.  The network-formation model and network statistics are set identified due to multiplicity of equilibria. I provide a computationally tractable outer approximation of the joint identified region for preferences determining network-formation processes and network statistics.  In a simulation study on Katz-Bonacich centrality, I find that worst-case bounds that do not use the network formation model are 44 times wider than the bounds I obtain from my procedure.

Working Papers

Discrete Choice under Risk with Limited Consideration, with Levon Barseghyan and Francesca Molinari.  R&R at American Economic Review.

This paper is concerned with learning decision makers' (DMs) preferences using data on observed choices from a finite set of risky alternatives with monetary outcomes. We propose a discrete choice model with unobserved heterogeneity in consideration sets (the collection of alternatives considered by DMs) and unobserved heterogeneity in standard risk aversion. In this framework, stochastic choice is driven both by different rankings of alternatives induced by unobserved heterogeneity in risk preferences and by different sets of alternatives considered. We obtain sufficient conditions for semi-nonparametric point identification of both the distribution of unobserved heterogeneity in preferences and the distribution of consideration sets. Our method yields an estimator that is easy to compute and that can be used in markets with a large number of alternatives. We apply our method to a dataset on property insurance purchases. We find that although households are on average strongly risk averse, they consider lower coverages more frequently than higher coverages. Finally, we estimate the monetary losses associated with limited consideration in our application. 

Calibrated Projection in MATLAB: Users’ Manual, with Hiroaki Kaido, Francesca Molinari, and Jörg Stoye.

 We present the calibrated-projection MATLAB package implementing the method to construct confidence intervals proposed by Kaido, Molinari and Stoye (2017). This manual provides details on how to use the package for inference on projections of partially identified parameters. It also explains how to use the MATLAB functions we developed to compute confidence intervals on solutions of nonlinear optimization problems with estimated constraints. 

Supporting code can be found here.   This program is distributed in the hope that it will be useful, but with any warranty; without even the implied warranty of merchantability or fitness for a particular purpose. If you use the software, we ask that you please cite Kaido, Molinari and Stoye (Econometrica, 2019) as the source of the theoretical results and of the code. 

Works in Progress

  • Estimating Peer Effects with a Dynamic Mismeasured Network.

Research Awards and Grants

  • L.R. “Red” Wilson MA ‘67 Excellence in Economics Medal, 2016

  • Tapan Mitra Economics Prize, 2016.

Room 433 Uris Hall 

Cornell University 

Ithaca, NY 14850 

t. (607) 262-9846



Stack Overflow