Grant: Discovery Grants Program - Individual
University of Victoria
$16,000.00
Hierarchical modeling and bayesian inference for spatial data
EF 2010-11
Résumé
The driving force behind my program of research lies in the development of statistical methods for modeling and inference with spatial and spatiotemporal data. These methods are applied with strong impact to answering scientific questions arising in both forestry and spatial epidemiology. In the forest ecology setting, spatiotemporal data arise in studies of recurrent tree infection where it is of interest to describe the
complex dynamics of disease as well its interaction with other processes such as wildfire. Understanding these dynamics can help mitigate the severe economic and environmental impact of such disease. In spatial epidemiology, it is of substantial interest to characterize variation in disease risk, both geographically and over
time in disease mapping studies. Here, statistical methods are developed with focus on modeling latent risk patterns and characterizing residual heterogeneity which can give clues on disease etiology and aid in generating related hypotheses.
The application areas generate challenging problems for statistical modeling, computing and inference. From a modeling perspective, there is a need to accommodate non-Gaussian, censored data with models allowing for flexible parametric or semiparametric dependence structure including spatial, temporal and more general multivariate correlation. From a computing perspective, there is a need to develop algorithms and approximate methods to handle the large spatial datasets that arise commonly in forestry and ecology. The research proposed here will tackle these problems with specific focus on the development of (I) methods for statistical modeling with large point-referenced spatial datasets and (II) flexible and robust models for disease mapping based on a new framework for Bayesian area-to-point kriging.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
British Columbia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
327131-2009
Name nom
Nathoo, Farouk
Department département
Mathematics and Statistics
Organization id
7
Institution établissement
University of Victoria
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2010
Competition year année de concours
2009
Award amount
16000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
14
Committee name en
Statistical Sciences
Committee name fr
Statistique
Area of application code
201
Area of application group en
Natural resources (economic aspects)
Area of application group fr
Ressources naturelles (aspects économiques)
Area of application en
Forestry (silviculture, forest management)
Area of application fr
Foresterie (sylviculture, aménagement forestier)
Research subject code
3004
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Applied statistics
Research subject fr
Statistique appliquée
Partie
2
Nb partie
5
Application title
Hierarchical modeling and bayesian inference for spatial data
Keyword
NSERC
Application summary
The driving force behind my program of research lies in the development of statistical methods for modeling and inference with spatial and spatiotemporal data. These methods are applied with strong impact to answering scientific questions arising in both forestry and spatial epidemiology. In the forest ecology setting, spatiotemporal data arise in studies of recurrent tree infection where it is of interest to describe the
complex dynamics of disease as well its interaction with other processes such as wildfire. Understanding these dynamics can help mitigate the severe economic and environmental impact of such disease. In spatial epidemiology, it is of substantial interest to characterize variation in disease risk, both geographically and over
time in disease mapping studies. Here, statistical methods are developed with focus on modeling latent risk patterns and characterizing residual heterogeneity which can give clues on disease etiology and aid in generating related hypotheses.
The application areas generate challenging problems for statistical modeling, computing and inference. From a modeling perspective, there is a need to accommodate non-Gaussian, censored data with models allowing for flexible parametric or semiparametric dependence structure including spatial, temporal and more general multivariate correlation. From a computing perspective, there is a need to develop algorithms and approximate methods to handle the large spatial datasets that arise commonly in forestry and ecology. The research proposed here will tackle these problems with specific focus on the development of (I) methods for statistical modeling with large point-referenced spatial datasets and (II) flexible and robust models for disease mapping based on a new framework for Bayesian area-to-point kriging.