Marc Comas-Cufí

Tenure-eligible lecturer in Statistics

Universitat de Girona

I am a member of the research group in Statistics and Compositional Data Analysis at the University of Girona. I enjoy programming and learning how to use new programming languages. For example, I program in R/Python for statistical tasks and data management, Java for GUI development and JavaScript when web programming is required. In the past, I have used C/C++, Fortran, Visual Basic (and VBA), Stata, Matlab, Maple and Lisp. I have good notions of most common markup languages: html, LaTeX or Markdown for documents and xml or json for data structures. In my daily work, I usually work with open-source software. Nowadays, the programs I use the most are RStudio, GNU Make and Git.


  • Multivariate Count Data analysis
  • Compositional Data modeling
  • Clustering methods
  • Competing Risk analysis
  • Causal inference
  • Survival analysis


  • PhD in Compositional Data Analysis, 2018

    Universitat de Girona

  • Master in High School Teaching, 2010

    Universitat de Girona

  • Master Thesis in Computer Science, 2007

    Universitat Politècnica de Catalunya

  • Master in Financial Engineering "Matemàtiques per als Instruments Financers", 2005

    Universitat Autònoma de Barcelona and Centre de Recerca Matemàtica

  • Degree in Mathematics, 2003

    Universitat Autònoma de Barcelona

Recent Publications


IFCS 2019

A log-ratio approach to cluster analysis of count data when the total is irrelevant


A compositional approach to the generalised propensity score


Could logratio methods be useful to analise the generalized propensity score?


Dealing with count zeros in compositional data analysis using the logratio-normal-multinomial distribution

SEIO 2018

Una aproximación composicional a la puntuación generalizada a la propensión a tratamiento


  • Campus Montilivi, P-IV Building, Office 255
    17071 Girona
  • Enter building P-IV and take the stairs to office 255 on last floor
  • Telegram Me