Research
My research sits at the intersection of artificial intelligence and statistical methodology, with a focus on survey sampling and missing data. I work across theory and practice, emphasizing tools that address selection bias, nonresponse, and the complexities of modern data sources. My contributions include generalized entropy calibration, imputation for high-dimensional data, and inference under complex sampling designs, as well as applied work integrating statistical methodology with artificial intelligence.
Research Areas
Survey Sampling & Missing Data
Calibration weighting, generalized entropy methods, model-assisted estimation, propensity score methods, non-probability samples
Transfer Learning
Calibration-based domain adaptation, distributional shift, multi-source data integration, diffusion model-based nonresponse adjustment
AI-Driven Data Analysis
High-dimensional variable selection, knockoff-based FDR control, ensemble methods, RNA-seq and genomic data analysis
Software
GECal — An R package for Generalized Entropy Calibration in survey sampling. Available on CRAN.
calibration — An R package providing an integrated workflow for survey-weight calibration with a modern API, structured solver diagnostics, and extensible solver backends.