United States
I am an economist and data scientist with a Ph.D. in Economics from Boston College. My work sits at the intersection of causal inference, structural modeling, experimentation, and AI, with a focus on using rigorous empirical methods to solve business and policy problems. I have built demand estimation and pricing tools at Amazon, supported litigation and damages analysis at The Brattle Group, and led research projects in affordable housing, health economics, and AI-integrated allocation design at Boston College. I work primarily in Python, SQL, and R, and I enjoy turning complex data into clear recommendations for product, strategy, and policy decisions. www.yunussemihcoskun.com
- Designed and taught courses in machine learning for economics, econometrics, and industrial organization, using Python, R, and Stata. - Received the Donald J. White Teaching Excellence Award (2024 and 2026) at Boston College. - Mentored students in empirical research, coding, and applied economic analysis. - Managed 10+ teaching assistants and graders for course coordination.
- Developed a policy analysis framework for affordable housing in Massachusetts by combining structural modeling, synthetic survey data, and mechanism design simulations to evaluate welfare and fairness under alternative participation costs and allocation methods. - Developed a synthetic survey methodology using large language models and ACS microdata to generate preference data at scale, and designed a validation experiment using Qualtrics. - Used shift-share instrumental variables to identify causal spillovers in Medicare markets and provider behavior. Built a county-level panel covering 3,000+ U.S. counties using ACS, BLS, and USDA data, and automated pipelines for cleaning, merging, and feature engineering.
- Supported research and undergraduate economics instruction through recitation leadership, grading, student guidance, and course coordination.
- Built an end-to-end SQL and AWS Athena pipeline for a large-scale demand estimation product supporting pricing and delivery decisions. - Estimated demand using discrete choice models in Python on AWS SageMaker and validated model performance against region-level experiments. - Ran counterfactual pricing simulations, recommended an optimized pricing, which projected multi-million-dollar revenue gains. - Presented findings to economists and senior stakeholders; authored a technical paper accepted at an internal Amazon research conference, and produced a business report contributing to the three-year growth strategy for a multi-billion-dollar product line.
- Applied difference-in-differences and double machine learning to large transaction panel data to estimate heterogeneous effects across customer subgroups and inform multimillion-dollar settlement amounts. - Built and stress-tested economic models to quantify large-scale consumer damages and support litigation strategy in antitrust matters. - Built reproducible analysis pipelines in R and supervised 10+ analysts across cases by delegating tasks, reviewing code, and enforcing quality and reproducibility standards. - Synthesized academic and case evidence to support expert testimony and produce client-ready materials.
- Co-taught Financial and Managerial Economics at Harvard Summer School, translating quantitative economic concepts into practical decision-making tools for students.
- Supported research and undergraduate economics instruction through recitation leadership, grading, student guidance, and course coordination.