Associate Instructor

E371 — Introduction to Applied Econometrics
Undergraduate Indiana University Bloomington SP 2025 · SP 2026 · FA 2026
This course introduces applied econometrics with emphasis on both implementation and reasoning. Students learn to translate economic questions into regression specifications, interpret results, and assess whether conclusions are credible. Core methods include linear regression, fixed effects for panel data, binary response models, and instrumental variables, all implemented in R using real datasets. AI tools are explicitly integrated into the course: students are encouraged to use them for coding, exploration, and drafting, while being held responsible for verifying correctness and defending every choice they make. The course concludes with an optional group mini-project in which students use AI to independently learn an econometric method and then teach it to the class with a live data illustration.
E370 — Statistical Analysis for Business and Economics
Undergraduate Indiana University Bloomington SU/FA 2023 · SP/SU 2024 · SU 2025 · SU 2026
This course introduces statistics as a science of understanding and analyzing economic data. Topics include graphical and numerical summaries of data, probability theory, population and sampling distributions, hypothesis testing, and regression analysis. Students develop the tools needed to answer real-world questions and gain a thorough understanding of statistical inference in economics and business contexts. Excel is used for data analysis and computation throughout the course.

Visiting Instructor

Structural Econometrics: Models and Applications
PhD Zhejiang University of Finance and Economics SU 2025
This course builds structural econometric thinking from the ground up. Rather than presenting models as a fixed toolkit, it follows a problem-driven progression: each model is motivated by a concrete economic question, its limitations are made explicit, and the next model is introduced as a direct response. Starting from the contrast between structural and reduced-form approaches, the course develops static discrete choice models — from logit to nested logit to random coefficients logit — by asking at each step what the current model cannot capture and why that matters for policy. It culminates in the Berry–Levinsohn–Pakes (BLP) framework, showing how random coefficients and market-level data together enable credible demand estimation in differentiated product markets. Implementation is central throughout: students estimate each model by hand in R and Python, working through the mechanics of MLE, GMM, and simulation-based methods on simulated data. The goal is not just to run models, but to develop the intuition to question them.