Grant: Discovery Grants Program - Individual
University of Manitoba
$15,000.00
Inference for discretely observed, continuous-time Markov processes on lattices
EF 2013-14
Résumé
Recent advances in digital recording technology have given researchers new ways of continually collecting data at many resolutions, both spatially and temporally. For example, Earth-orbiting satellites are increasingly used to capture environmental data, such as precipitation, polar sea ice concentrations and sea surface temperatures. The observed processes can be viewed as dynamical systems that can be further characterized locally in space and time. Population dynamics models are often specified in continuous time; however, it is usually the case that the processes are sampled at discrete time points, so that state sequences are not fully observed. Thus, one objective of the proposed research program will be the development of statistical methodologies for discretely observed spatial population data by treating the underlying systems as continuous-time Markov processes. In particular, I will concentrate on Interacting Particle Systems (IPS) and spatial Birth-death-migration (BDM) processes, with applications in environmetrics (sea ice data), biology (tumor growth data), ecology (population growth data), and genetics (genome activation data). IPSs, a subject of interest in probability theory, has received little attention in statistics, partly due to the fact that the necessary equations are mathematically intractable. However, given current advances in computational methods, there has been a resurgence of interest in developing approximate solutions for more complex and more appropriate models of population dynamics. Thus, another objective of the program is the study of IPSs from statistical perspectives. The anticipated outcomes of the research program are the development of inferential procedures for IPSs and BDMs, as well as their application to real world problems. The research program will focus on the following topics: (1) numerical approximation of transition probabilities and of conditional moments; (2) inference and efficient estimation of structural parameters; and (3) goodness-of-fit measures.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Manitoba, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
418368-2013
Name nom
Koulis, Theodoro
Department département
Statistics
Organization id
19
Institution établissement
University of Manitoba
Province en
Manitoba
Province fr
Manitoba
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2013
Competition year année de concours
2013
Award amount
15000
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
1205
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Mathematical sciences
Area of application fr
Sciences mathématiques
Research subject code
3007
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Stochastic processes
Research subject fr
Processus stochastiques
Partie
1
Nb partie
5
Application title
Inference for discretely observed, continuous-time Markov processes on lattices
Keyword
NSERC
Application summary
Recent advances in digital recording technology have given researchers new ways of continually collecting data at many resolutions, both spatially and temporally. For example, Earth-orbiting satellites are increasingly used to capture environmental data, such as precipitation, polar sea ice concentrations and sea surface temperatures. The observed processes can be viewed as dynamical systems that can be further characterized locally in space and time. Population dynamics models are often specified in continuous time; however, it is usually the case that the processes are sampled at discrete time points, so that state sequences are not fully observed. Thus, one objective of the proposed research program will be the development of statistical methodologies for discretely observed spatial population data by treating the underlying systems as continuous-time Markov processes. In particular, I will concentrate on Interacting Particle Systems (IPS) and spatial Birth-death-migration (BDM) processes, with applications in environmetrics (sea ice data), biology (tumor growth data), ecology (population growth data), and genetics (genome activation data). IPSs, a subject of interest in probability theory, has received little attention in statistics, partly due to the fact that the necessary equations are mathematically intractable. However, given current advances in computational methods, there has been a resurgence of interest in developing approximate solutions for more complex and more appropriate models of population dynamics. Thus, another objective of the program is the study of IPSs from statistical perspectives. The anticipated outcomes of the research program are the development of inferential procedures for IPSs and BDMs, as well as their application to real world problems. The research program will focus on the following topics: (1) numerical approximation of transition probabilities and of conditional moments; (2) inference and efficient estimation of structural parameters; and (3) goodness-of-fit measures.