Igor Geyn — Data Scientist & Policy Analyst
I specialize in applying causal inference methods to policy and business challenges in the energy sector. Additionally, my research includes applications of machine learning and Bayesian inference/analysis.
At PG&E, I use modern econometric techniques—including difference in differences (DiD) and regression discontinuity (RD)—to evaluate utility programs and inform enterprise risk management decisions. I combine large-scale data analysis with rigorous statistical methods to study how policies and programs affect real-world outcomes. My research has been published in the American Political Science Review and is forthcoming in the American Economic Journal: Economic Policy.
Through direct employment and consulting engagements—primarily through Atlantic Media—I have experience in nearly a dozen distinct verticals and have consulted with several dozen different companies spanning the health sector, defense and civilian federal contracting, IT modernization and other technology verticals, and state and local government.
Current Work
I'm a data scientist at Pacific Gas & Electric (PG&E), where I use advanced econometric methods to evaluate company programs and inform risk management strategies. My work involves applying the same rigorous causal inference techniques used in top academic research to real-world policy and operational challenges.
I'm also a PhD candidate in Political Science at UCLA (focused on American politics and methods/political methodology), where I developed expertise in experimental design, statistical analysis, policy evaluation, and assessing/using game theoretic models of behavior. I am working towards completing my dissertation by winter 2026.
Recent Research
My academic work has been published in leading journals including the American Political Science Review and is forthcoming in the American Economic Journal: Economic Policy. At a high level, this research applies rigorous quantitative methods to understand the effects of institutions and policies on democratic participation and governance. More specifically, my published work:
analyzes the effects of local partisan election administration in the U.S. (published at the American Political Science Review; co-authored with Joshua Ferrer and Dan Thompson),
provides causal evidence of the effect of higher education on political participation (conditionally accepted at the American Economic Journal: Economic Policy; co-authored with Daniel Firoozi).
My work in progress examines the effect of political institutions (e.g., ballot propositions) on voters' ability to learn about policy as well as the effect of partisan alignment between members of Congress and the President on disaster relief outcomes.
Past analytical work includes policy publications on regional disparities in the behavioral health workforce, forecasts of primary care supply and demand in California, and workforce analyses of health providers in Washington, D.C.; work published through UCSF and the D.C. Policy Center, respectively.
Let's Connect!
Contact: igorgeyn at gmail dot com
CV: [link]