Sima Khoei

I’m a biostatistics Ph.D. student dedicated to building reliable, data-driven tools for infectious-disease modeling and public-health decision-making. My current work focuses on machine-learning–based calibration for agent-based epidemic models—including a three-layer bidirectional LSTM calibrator that improves accuracy and runtime over ABC and is being productized as epiworldRcalibrate—and on the comparative evaluation of effective reproduction number (R_t) estimators using agent-based network models. I collaborate closely with Dr. Bernardo Modenesi, Dr. Yue Zhang, and Dr. George Vega Yon, and I contribute to open-source tools such as epiworldR.

Beyond methods, I develop R packages with clear APIs, careful documentation, and reproducible workflows. Current efforts include epiworldRcalibrate for practical calibration of epidemic ABMs and imaginarycss for Cognitive Social Structures (software paper in preparation, Sept. 2025). I share results with both technical and applied audiences (CDC 2024; ENAR 2025; JSM 2025 selection) with the goal of turning rigorous statistics and machine learning into actionable public-health insight.