Homepage Thomas Opitz

Research Associate (CR)

228, Route de l'Aérodrome
CS 40509
84 914 Avignon Cedex 9

Telephone: 04 32 72 21 86
Mail: thomas POINT opitz AT inra POINT fr


Principal research interests

I am  interested in stochastic geometry and in spatial and spatiotemporal statistics with a view towards extreme values, often with a blend of frequentist and Bayesian inference techniques.

  • Applications of Integrated Nested Laplace Approximation to spatially structured Bayesian hierarchical models
  • Lévy-based spatial modeling beyond Gaussian random fields
  • Spatial extreme value modeling between asymptotic dependence and independence
  • Modeling climatic processes and the associated extreme risks
  • Spatial modeling of the occurrence of forest fires and landslide events
  • Spatial and spatiotemporal modeling of plant and animal species

Ongoing projects

  • Steering committee of RESSTE ("RESeau Statistique pour données Spatio-TEmporelles"), one of INRA's current research groups
  • Member of LEFE-CERISE
    "Simulation de scénarii intégrant des champs extrêmes spatio-temporelle avec éventuelle indépendance asymptotique pour des études d'impact en science de l'environnement"

Current work in progress

  • [with F. Bonneu, E. Gabriel] Statistical space-time analysis of wildfires in Southern France based on point process methods
  • [with L. Lombardo, R. Huser] Spatial modeling of landslide events with log-Gaussian Cox processes
  • [with A. Schoeny, J. Papaix] Modeling the climatic effects on aphid species dynamics in melon fields
  • [with J.N. Bacro, C. Gaetan, G. Toulemonde] Multivariate and space-time generalized Pareto models for threshold exceedances based on exponential-gamma ratios
  • [with R. Huser, E. Thibaud] Max-infinitely divisible models for subasymptotic modeling of spatial maxima data
  • INLA-based nonparametric statistical inference of space-time trends in mean, variance and tail behavior of global monthly temperature data
  • Lévy moving averages with indicator kernels: Construction and inference for random field models beyond the Gaussian
  • Higher-order Markov models for spatially indexed  time series with local distributions parametrized through space-time covariances


  1. Jean-Noel Bacro, Carlo Gaetan, Thomas Opitz, and Gwladys Toulemonde. Hierarchical space-time modeling of exceedances with an application to rainfall data. Submitted, 2017.
    Luigi Lombardo, Thomas Opitz, and Raphael Huser. Point process-based modeling of multiple debris flow landslides using inla: an application to the 2009 messina disaster. Submitted, 2017.
    Gabriel, E., Opitz, T., Bonneu, F.  Detecting and modeling multi-scale space-time structures: the case of wildfire occurrences. Submitted to Journal de la Société Française de Statistique (Special Issue on Space-Time Statistics). [pdf link]
    RESSTE Network. Analyzing spatio-temporal data with R: Everything you always wanted to know -- but were afraid to ask. Submitted to Journal de la Société Française de Statistique (Special Issue on Space-Time Statistics). [pdf link]



Bayesian spatial modeling

Opitz, T. Latent Gaussian modeling and INLA: A review with focus on space-time applications. To appear in Journal of the French Statistical Society (Special Issue on Space-Time Statistics). [pdf link]

Extreme value analysis and risk

Huser, R., Opitz, T., Thibaud, E. Bridging Asymptotic Independence and Dependence in Spatial Extremes Using Gaussian Scale Mixtures . Spatial Statistics 21, 166–186 (2017). [arXiv]
Mornet, A., Opitz, T., Luzi, M., Loisel, S. & Bailleul, B. Wind Storm Risk Management: Sensitivity of Return Period Calculations and Spread on the Territory. Accepted for Stochastic Environmental Research and Risk Assessment.
Opitz, T. Modeling asymptotically independent spatial extremes based on Laplace random fields. Spatial Statistics 16, 1–18 (2016). [arXiv]
Thibaud, E. & Opitz, T. Efficient inference and simulation for elliptical Pareto processes. Biometrika 102(4), 855–870 (2015).
Mornet, A., Opitz, T., Luzi, M. & Loisel, S. Index for Predicting Insurance Claims from Wind Storms with an Application in France. Risk Analysis 35(11), 2029–2056 (2015).
Opitz, T., Bacro, J.-N. & Ribereau, P. The spectrogram: A threshold-based inferential tool for extremes of stochastic processes. Electronic Journal of Statistics 9, 842–868 (2015). [open access link]
Opitz, T. Extremal processes: Elliptical domain of attraction and a spectral representation. Journal of Multivariate Analysis 122, 409–413 (2013). [arXiv]

Text mining

Donald Tapi Nzali, Mike and Azé, Jérôme and Bringay, Sandra and Lavergne, Christian and Mollevi, Caroline and Opitz, Thomas. What Patients Can Tell Us: Topic Analysis for Social Media on Breast Cancer. Journal of Medical Internet Research, 2017.
Donald Tapi Nzali, M., Azé, J., Bringay, S., Lavergne, C., Mollevi, C. & Opitz, T. Formalisation semi-automatique d'un vocabulaire Patient/Médecin dédié au cancer du sein. Revue d'Intelligence Artificielle, Numéro spécial IC 2014/2015.
Opitz, T. et al. Breast cancer and quality of life: medical information extraction from health forums. in Medical Informatics Europe, pages 1070–1074 (2014).
Opitz, T. et al. Paroles de patients dans les forums de santé: une perspective originale sur la qualité de la vie. in 25es Journées francophones d’Ingénierie des Connaissances, Actes de l’Atelier IA & Santé (2014).


Opitz, T. Extrêmes multivariés et spatiaux: approches spectrales et modèles elliptiques. (PhD Thesis, University of Montpellier, 2013).
Opitz, T., Tramini, P. & Molinari, N. Spline regression for zero-inflated models. JP Journal of Biostatistics 9(1), 53–66 (2013).

Current responsibilities

  • Webmaster (web site and registration, submission, review systems) of the METMA IX conference on Space-time modeling and statistics, Summer 2018, Montpellier
  • Co-organizer of the weekly seminar of BioSP

Past responsibilities

  • Webmaster (web site and registration, submission, review systems) of the Journées de Statistique 2017, Avignon
  • Organization and Program Committee of AG MIA 2017 / Journées Maths-Infos / Journées INRA-Inria
  • Program Committee of Spatial Statistics 2015, Avignon


  • 2016/2017: Statistique Descriptive 2, L1 STID, IUT Avignon