I am Geonhee Han (韓 健熙). You can call me Geonhee, GH, Geo, or Han. I am currently a first-year Ph.D. student at Columbia Statistics. My research interests are Bayesian statistics and statistical computing. I am very fortunate to have been mentored by Kaoru Irie (at UTokyo) and Andrew Gelman (at Columbia), who shaped my path into the field from a different background. I buy books faster than I read them. Feel free to reach out if you'd like to talk.
Zero-inflated stochastic volatility model for disaggregated inflation data with exact zeros
Han, G. & Irie, K. (2026+)
Journal of the Royal Statistical Society, Series C (Applied Statistics), qlag026.
Adaptive sequential Monte Carlo for structured cross validation in Bayesian hierarchical models
Han, G. & Gelman, A. (2026)
Journal of Computational and Graphical Statistics, 35(3).
Gradient-free Gibbs sampler for shallow-wide Bayesian neural networks
Han, G. (2025)
Statistics and Computing, 35(4), 111.
Grappling with uncertainty in forecasting the 2024 US presidential election
Gelman, A., Goodrich, B. & Han, G. (2024)
Harvard Data Science Review, 6(4).
Efficient scenario analysis in real-time Bayesian election forecasting via sequential meta-posterior sampling
Han, G., Gelman, A. & Vehtari, A. (2025)
under revision. 2510.19133 [stat.ME]