I am an economist and data scientist specializing in causal inference, econometrics, experimentation, and applied machine learning. My work focuses on moving from observed relationships to credible causal questions—and translating quantitative evidence into decisions.
A series on modern causal inference, identification, and the practical interpretation of causal estimates.
Staggered treatment adoption, heterogeneous treatment effects, and why the familiar TWFE coefficient may not identify the causal parameter we have in mind.
Read Article →Group-time treatment effects, dynamic effects, and why defining the estimand must come before choosing the estimator.
Read Article →Event-time effects, contaminated comparisons, pre-trend testing, and why a convincing graph cannot substitute for identification.
Read Article →Moving from ATE to CATE using causal machine learning, causal forests, cross-fitting, and heterogeneous treatment effects.
Read Article →Reproducible empirical projects connecting modern econometric methods with real data and decision problems.
An empirical causal-ML workflow comparing raw associations, regression adjustment, IPW, AIPW, Causal Forest DML, and T-learners to study heterogeneous effects on household net financial assets.
View Project →A reproducible analysis of staggered treatment adoption, comparing conventional TWFE event studies with modern group-time treatment-effect estimators.
View Project →My doctoral research examines education, fertility, female labor force participation, and gender wage disparities using large-scale microdata and applied econometric methods.
This chapter studies the causal effect of women's education on fertility outcomes during Iran's dramatic fertility transition. I construct a new dataset on elementary and secondary school expansion and combine it with multiple waves of national census microdata to address the endogeneity of educational attainment.
Methods: Instrumental Variables · Large-Scale Census Microdata · Applied Econometrics
Read Chapter Summary →This chapter examines how higher education and fertility relate to women's labor force participation using repeated cross-sectional census data. The empirical strategy addresses the endogeneity of education and fertility using separate instrumental-variable approaches.
Methods: Instrumental Variables · Repeated Cross Sections · Labor Economics
Read Chapter Summary →This chapter examines gender wage disparities using household income and expenditure microdata spanning 1994–2019. It studies how the magnitude and evolution of wage differences vary across occupations, industries, education levels, and family characteristics.
Methods: Wage Analysis · Microdata · Distributional Analysis · Labor Economics
Read Chapter Summary →
I hold a PhD in Economics from a U.S. university, with additional
academic training in Economics, Applied Mathematics, and Industrial
Engineering.
I currently work at the University of California, Berkeley.
My quantitative interests include causal inference, econometrics,
experimentation, statistical modeling, and applied machine learning.