Assistant Professor of Economics, University of Pennsylvania
Department of Economics
University of Pennsylvania
3718 Locust Walk
Philadelphia, PA 19104
Curriculum Vitae pdf
I am on leave at the Cowles foundation for the academic year 2017-2018.
Kernel Filtering pdf, Github code, offline code
The paper describes a new approximate nonlinear filtering technique. Strengths of the technique include: (1) it can be used as long as one can simulate from the model, without the need to evaluate measurement densities, (2) it is easy to implement, yet competitive with state-of-the-art alternative techniques in terms of speed and accuracy, (3) it can be used with some models that include infinite-dimensional state variables. The main theoretical result of the paper is that the approximation error of the technique goes to zero with computational power, and that it does so uniformly with respect to the time horizon of the data.
Hidden Rust Models pdf, supplement pdf
Hidden Rust models answer the need for convenient models of unobserved dynamics in a dynamic discrete choice context. They can be thought of as classical Rust models with an additional unobserved state variable. As opposed to their classical counterparts, hidden Rust models are not Markovian. I study their econometrics in terms of: identification, time-series asymptotics, practical estimation. I illustrate their use in a model of dynamic financial incentives inspired by Duflo, Hanna and Ryan (2012). Several lessons carry over to other structural models with unobserved dynamics.
A Weakly Dependent Bernstein–von Mises Theorem pdf
The Bernstein–von Mises theorem is the central result of classical asymptotic theory applied to Bayesian methods. I isolate a convenient set of assumptions and prove a new weakly dependent Bernstein–von Mises theorem along the lines of Le Cam (1986). This new theorem is valid under few assumptions beyond local asymptotic normality. An application in a microeconomic model of dynamic choice can be found in my job market paper.