Coursework

Demand. Random-coefficients logit with optimal instruments, diversion ratios and merger counterfactuals, estimated with pyBLP [Notebook] [Code] [Write-up]

Single-agent dynamics. Value-function iteration and structural estimation of a Rust-style machine-replacement model on course data [Notebook]

Dynamic games. Bajari–Benkard–Levin forward simulation with conditional choice probabilities, recovering profit parameters from an entry/exit game [Code] [Write-up]

Four write-ups: matching estimators and treatment effects; martingales, uniform integrability and tail sigma-algebras; pre-testing and the uniformity of inference; the Wold decomposition; stable convergence and Bartik instruments; limit theory for dyadic data and under two-way dependence [1] [2] [3] [4]

Two longer course requirements: a class presentation of Andrews, Barahona, Gentzkow, Rambachan and Shapiro, “Structural Estimation Under Misspecification: Theory and Implications for Practice” (QJE 140(3), 2025) [Presentation], and a referee report on Borusyak, Chen, Hull and Lei, “Nonparametric Identification of Demand without Exogenous Product Characteristics” (NBER WP 34842) [Referee report]

Five write-ups: parametric and nonparametric identification; asymptotic theory for nonlinear models; limited-dependent-variable models; quantile instrumental variables; Kolmogorov–Smirnov and related distributional tests; the bootstrap and subsampling [1] [2] [3] [4] [5]