Diego Serrano Ortega

I am a PhD candidate in Statistics for Data Science at the Department of Statistics of Universidad Carlos III de Madrid, and a member of the NICDA research group (Nonparametric Inference for Complex Data and its Applications). My thesis, Topics on random forests in metric spaces, is supervised by Eduardo García-Portugués and Marco Raiola.

I read Mathematics at Universidad Autónoma de Madrid (BSc, 2018–2022) and Statistics for Data Science at Universidad Carlos III de Madrid (MSc, 2022–2023). I have devoted myself to research and teaching since October 2023.

Research

I work on inference for random objects: data valued in a general metric space rather than a Euclidean one — distributions, covariance matrices, curves, directions, shapes. Only a distance is assumed, so the tools apply across very different kinds of data, at the price of having to redefine very basic notions such as the mean or the variance. My thesis envisages an application of metric space data analysis to aeroacoustics, since the data used in this area — flow fields and acoustic curves — is often complex.

In the first work of my thesis we introduced out-of-bag prediction balls: uncertainty regions for the prediction of any bagging regression model with responses in a metric space, with asymptotic coverage guarantees. More recently, together with Ingrid Van Keilegom (KU Leuven) and Eduardo García-Portugués (UC3M), I have been working on a goodness-of-fit framework for distributions on separable metric spaces, built on distance profiles.

Research interests

  • Statistics for random objects and metric-space-valued data
  • Goodness-of-fit testing
  • Empirical processes
  • Nonparametric statistics
  • Uncertainty quantification and prediction regions
  • Random forests and Fréchet regression

Education

PhD in Statistics for Data Science · since November 2024
Universidad Carlos III de Madrid. Thesis: Topics on random forests in metric spaces. Supervisors: Eduardo García-Portugués (Department of Statistics) and Marco Raiola (Department of Aerospace Engineering). Expected 2028.
MSc in Statistics for Data Science · 2022–2023
Universidad Carlos III de Madrid. Best academic record of the graduating class. Thesis: Shape-constrained estimation in functional regression by means of P-splines (supervisors: M. Durbán and V. Guerrero).
BSc in Mathematics · 2018–2022
Universidad Autónoma de Madrid. Thesis: Continued fractions and some infinite sums (supervisor: F. Chamizo).

Awards and honours

  • Best Poster Award
  • First prize, Thesis Talk 2025
  • Extraordinary Master's Degree Award
  • First prize, Master's Excellence Awards Programme

Research stays

  • ORSTAT, KU Leuven, Belgium

Funding and contracts

  • Predoctoral research contract (PIPF)
  • EuroCC Spain Testbed award
  • Team member, grant PID2024-158399NB-I00
  • Research technician
  • Mobility grants for the KU Leuven visits