Grant: Discovery Grants Program - Individual
McGill University
$35,000.00
Functional data analysis for spatial and dynamic data
EF 2013-14
Résumé
Functional data deals with data distributed over continua of various sorts, of which time is the best known example; and is distinguished from the many other areas in statistics that consider such data by avoiding restrictive assumptions, such as equally-spaced measurements and stationarity; while at the same time often assuming smoothness and consequently that the data will support the estimation of a certain number of derivatives. Ramsay and Silverman (2005) remains the primary reference to functional data analysis, but their treatment is limited to considering data distributed over a single argument dimension, which is usually time. In collaboration with Dr. Jason Nielsen, Carleton University, Dr. Timothy Ramsay, Ottawa Hospital Research Institute and Dr. Laura Sangalli, Politecnico Milano, I will extend functional data analysis to deal with data over complex spatial and spatial/temporal regions, including regions with complicated exterior and interior boundaries. A textbook on spatial data analysis with finite elements is in the advanced planning stages, and will be written over the proposed grant period.A second development from functional data analysis is the use of systems of differential equations, or dynamic systems, to model functional data, and a major advance in this direction was taken in the JRSS-B discussion paper by Ramsay, Hooker, Campbell and Cao (2007) who developed an effective parameter estimation strategy for models of this type. Hooker and I have signed a publishing contract with Springer, and will write a text for the statistical community building on this work, software in Matlab and R, and the already extensive range of applications that we and others have achieved. The two programs will interact in terms of research on parameter estimation for partial differential equation systems, since PDE's are a powerful tool for modeling the spatial and spatial/temporal structure of spatial data.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
320-2012
Name nom
Ramsay, James
Department département
Psychology
Organization id
61
Institution établissement
McGill University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2013
Competition year année de concours
2012
Award amount
35000
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
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et statistique
Area of application code
1200
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Advancement of knowledge
Area of application fr
Avancement des connaissances
Research subject code
3000
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Statistics and probability
Research subject fr
Statistique et probabilité
Partie
2
Nb partie
5
Application title
Functional data analysis for spatial and dynamic data
Keyword
NSERC
Application summary
Functional data deals with data distributed over continua of various sorts, of which time is the best known example; and is distinguished from the many other areas in statistics that consider such data by avoiding restrictive assumptions, such as equally-spaced measurements and stationarity; while at the same time often assuming smoothness and consequently that the data will support the estimation of a certain number of derivatives. Ramsay and Silverman (2005) remains the primary reference to functional data analysis, but their treatment is limited to considering data distributed over a single argument dimension, which is usually time. In collaboration with Dr. Jason Nielsen, Carleton University, Dr. Timothy Ramsay, Ottawa Hospital Research Institute and Dr. Laura Sangalli, Politecnico Milano, I will extend functional data analysis to deal with data over complex spatial and spatial/temporal regions, including regions with complicated exterior and interior boundaries. A textbook on spatial data analysis with finite elements is in the advanced planning stages, and will be written over the proposed grant period.A second development from functional data analysis is the use of systems of differential equations, or dynamic systems, to model functional data, and a major advance in this direction was taken in the JRSS-B discussion paper by Ramsay, Hooker, Campbell and Cao (2007) who developed an effective parameter estimation strategy for models of this type. Hooker and I have signed a publishing contract with Springer, and will write a text for the statistical community building on this work, software in Matlab and R, and the already extensive range of applications that we and others have achieved. The two programs will interact in terms of research on parameter estimation for partial differential equation systems, since PDE's are a powerful tool for modeling the spatial and spatial/temporal structure of spatial data.