CoDaPack
Stand-alone, multiplatform software for compositional data analysis with an accessible graphical interface.
Professional Summary
I am an Associate Professor in Statistics at the University of Girona and a member of the Statistics and Compositional Data Analysis research group.
My research connects statistical methodology with applied collaborations in health, epidemiology, environmental data, and clinical research. I work mainly on compositional and multivariate count data, clustering, survival analysis, competing risks, causal inference, and reproducible statistical workflows.
I teach statistics in undergraduate degrees and machine learning in a Master’s programme in Data Science. I enjoy bringing statistical ideas close to real data problems, both in teaching and in collaborative research.
I am open to consulting and collaborative projects involving statistical modelling, health data analysis, reproducible workflows, and methodological support for applied research.
Selected Publications
(2025).
Absolute impact of lipid levels on coronary heart disease in a real-world primary prevention cohort of 730,236 people.
American Journal of Preventive Cardiology, Online ahead of print.
(2025).
Four-year respiratory consequences of COVID-19 related pneumonia: a longitudinal cohort study.
BMC Pulmonary Medicine, 25:302.
(2020).
Modelling count data using the logratio-normal-multinomial distribution.
Statistics and Operations Research Transactions, 44(1):99-126.
(2019).
Epidemiology of dementia: prevalence and incidence estimates using validated electronic health records from primary care.
Clinical Epidemiology, 11:217-228.
Software
Open-source tools for compositional data analysis.
coda.base
Core R package for compositional data analysis using the log-ratio approach.
coda.plot
R package for visualising compositional data and log-ratio coordinates.
coda.pack
Meta-package providing a single entry point to the main stable packages in the coda ecosystem.
coda.count
Development R package for the analysis of compositional count data.
Selected Talks
Gaussian zero replacement method for compositional count data sets
Gaussian zero replacement method for compositional count data sets
Model-based zero replacement method for compositional count data sets
Model-based zero replacement method for compositional count data sets
Clustering count data using compositional methods
Clustering count data using compositional methods
A new approach for approximating the parameters of a log-ratio-normal-multinomial distribution
A new approach for approximating the parameters of a log-ratio-normal-multinomial distribution
Clustering count data using compositional methods
Clustering count data using compositional methods
Contact
For consulting, research collaborations, or methodological support, feel free to get in touch.
Universitat de Girona
Campus Montilivi, P-IV Building, Office 255
17071 Girona