F Nous contacter

Multi angle

H 1 High-dimensional Bayesian geostatistics ​

Auteurs : Banerjee, Sudipto (Auteur de la Conférence)
CIRM (Editeur )

    Loading the player...

    Résumé : With the growing capabilities of Geographic Information Systems (GIS) and user-friendly software, statisticians today routinely encounter geographically referenced data containing observations from a large number of spatial locations and time points. Over the last decade, hierarchical spatiotemporal process models have become widely deployed statistical tools for researchers to better understand the complex nature of spatial and temporal variability. However, fitting hierarchical spatiotemporal models often involves expensive matrix computations with complexity increasing in cubic order for the number of spatial locations and temporal points. This renders such models unfeasible for large data sets. I will present a focused review of two methods for constructing well-defined highly scalable spatiotemporal stochastic processes. Both these processes can be used as ``priors" for spatiotemporal random fields. The first approach constructs a low-rank process operating on a lower-dimensional subspace. The second approach constructs a Nearest-Neighbor Gaussian Process (NNGP) that ensures sparse precision matrices for its finite realizations. Both processes can be exploited as a scalable prior embedded within a rich hierarchical modeling framework to deliver full Bayesian inference. These approaches can be described as model-based solutions for big spatiotemporal datasets. The models ensure that the algorithmic complexity has n floating point operations (flops), where n is the number of spatial locations (per iteration). We compare these methods and provide some insight into their methodological underpinnings.

    Keywords : Bayesian modeling; directed acyclic graphs; Gaussian processes; low-rank models; scalable models; spatial stochastic processes

    Codes MSC :
    62F15 - Bayesian inference
    62M30 - Statistics of spatial processes
    62P12 - Applications of statistics to environnemental and related topics

      Informations sur la Vidéo

      Réalisateur : Hennenfent, Guillaume
      Langue : Anglais
      Date de publication : 06/12/2018
      Date de captation : 29/11/2018
      Collection : Research talks ; Probability and Statistics
      Format : MP4
      Durée : 01:00:45
      Domaine : Probability & Statistics
      Audience : Chercheurs ; Doctorants , Post - Doctorants
      Download : https://videos.cirm-math.fr/2018-11-29_Banerjee.mp4

    Informations sur la rencontre

    Nom de la rencontre : Jean-Morlet chair: Bayesian statistics in the big data era / Chaire Jean-Morlet : Statistiques bayésiennes à l'ère du big data
    Organisateurs de la rencontre : Freyemurth, Jean-Marc ; Marin, Jean-Michel ; Mengersen, Kerrie ; Pommeret, Denys ; Pudlo, Pierre
    Dates : 26/11/2018 - 30/11/2018
    Année de la rencontre : 2018
    URL Congrès : https://www.chairejeanmorlet.com/2018-2-...

    Citation Data

    DOI : 10.24350/CIRM.V.19478503
    Cite this video as: Banerjee, Sudipto (2018). High-dimensional Bayesian geostatistics ​. CIRM. Audiovisual resource. doi:10.24350/CIRM.V.19478503
    URI : http://dx.doi.org/10.24350/CIRM.V.19478503

    Voir aussi


    1. Banerjee, S. (2017). High-dimensional Bayesian geostatistics. Bayesian Analysis, 12(2), 583-614 - https://doi.org/10.1214/17-BA1056R