Research

* denotes equal contribution.

Accepted Publications

Li, M.*, Han, K.*, & Ma, Y. From Matrix Inversion to Constraints: Provably Tighter Confidence Regions for Importance Weights in Label Shift. Accepted at Neural Information Processing Systems (NeurIPS) 2026. [arXiv]

Under Review

Han, K., Ma, Y., Marder, K., & Garcia, T. (2026). SPYCE: A Doubly Robust Estimator for Trials Targeting Early Huntington Disease under Outcome-Dependent Censoring. [arXiv]

Han, K., Ma, Y., Marder, K., & Garcia, T. (2026). Semiparametric Prediction with Efficient Interval Length under a Right-Censored Covariate. [arXiv]

Working Papers

Han, K., Ma, Y., Marder, K., & Garcia, T. Nonparametric Monotonicity-Preserving Quantile Regression under a Right-Censored Covariate.

Han, K.*, Li, M.*, & Ma, Y. Quantile Optimization for Honest Conformal Prediction under Label Shift.

Han, K., & McCartan, C. Prediction Sequence: Anytime-Valid Regression-Based Sequence of Prediction Intervals.

Software

spyce — Estimating a regression parameter in a right-censored covariate setting under outcome-dependent censoring.

prescco — Producing a prediction interval under a right-censored covariate.

Presentations

Penn State SMAC Talk, University Park, PA (September 2026). Prediction Sequence: Anytime-Valid Regression-Based Sequence of Prediction Intervals. Department of Statistics, Penn State University.

2026 Joint Statistical Meeting, Boston, MA (August 2026). Semiparametric Prediction under Right-Censored Covariates. Improving and Assessing Risk Prediction Models session.

2025 Joint Statistical Meeting, Nashville, TN (August 2025). SPYCE: Semi-Parametric Y-dependent right-Censored Covariate Estimator. Semiparametric Modeling Advanced Research session.