The Nonparametric Inference for Complex Data and its Applications (NICDA) research group is formed by members of the Department of Statistics of Universidad Carlos III de Madrid. Its research focus is on nonparametric, mathematical, and computational statistics, with particular emphasis on the analysis of complex data with geometric structure, spatio-temporal dependence, or incomplete information.
The next book to be read in the Statistics Reading Group will be Computer Age Statistical Inference: Algorithms, Evidence and Data Science by Bradley Efron and Trevor Hastie. The book delivers a masterful synthesis of classical statistical inference and modern computational methods, tracing their evolution over the past 60 years. Insightful, rigorous, and historically grounded, it is an essential reading to understand the theoretical foundations and algorithmic developments shaping today’s data science.
Diego Serrano has obtained the best poster award in the International Workshop on Functional and Operatorial Statistics (IWFOS) held in Novara (Italy) from 25th to 27th June 2025. IWFOS is the leading international conference on functional data analysis.
The paper “A family of toroidal diffusions with exact likelihood inference”, coauthored by Eduardo García-Portugués and Michael Sørensen, has been accepted for publication in the renowned Biometrika! The preprint can be checked on the arXiv.
Diego Serrano has obtained the first award in the Thesis Talk 2025 contest organized by the Universidad Carlos III de Madrid. The Thesis Talk is a challenging competition for PhD students to present their research in a 4-minute talk to a general audience, with the goal of making complex topics accessible and engaging.
We are looking for motivated and talented individuals to join us in researching cutting-edge topics in mathematical and computational statistics for the analysis of complex data.
We welcome applications for the positions of:
If you are interested in joining our research group, please send your expression of interest through this form.