Rosalba Radice

Rosalba Radice is a Professor in the Department of Mathematics within the School of Mathematics, Statistics and Actuarial Science at City, University of London. With over 40 publications spanning more than a decade, she has established herself as a leading researcher in advanced statistical methodology with significant applications in health economics and epidemiology. Her work bridges theoretical statistics with practical policy applications, particularly in healthcare evaluation and economic analysis.

Radice's research focuses on developing sophisticated statistical models to address complex data challenges. She specializes in copula regression techniques, additive models, and methods for handling missing data and endogeneity. Her work provides innovative solutions for analyzing multivariate outcomes, time-to-event data, and complex survey designs. She has made significant contributions to distributional analysis, survival modeling, and causal inference methods that account for unobserved confounding.

Her publication record reveals a strong emphasis on applying advanced statistical methods to pressing health economics questions. Recent work examines healthcare utilization patterns, mental health spending effects, hospital-acquired infections, and the impact of health insurance provisions. She frequently collaborates with researchers across disciplines, demonstrating the versatility of her methodological contributions. Her articles consistently appear in top-tier journals across statistics, health economics, and epidemiology.

Radice has developed several important methodological frameworks, including copula-based regression models with flexible margins, switching regime regression approaches, and spline-based frameworks for multistate survival processes. Her work on addressing non-ignorable missing data using copula selection models has been particularly influential in health econometric evaluation. She has also made significant contributions to robust fitting procedures for generalized additive models and methods for analyzing bivariate time-to-event outcomes.