Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Grants Program - Individual

University of Alberta

$16,000.00

Finite mixture models and their applications

EF 2010-11

Résumé
Finite mixture models are widely applied to data sets in which several homogeneous populations are mixed together. In financial economics, mixtures of three normal distributions are often applied to daily stock returns due to the existence of three risk regimes: low, normal and high. The number of homogeneous populations in a finite mixture is of great interest in applications. For example in statistical genetics, this number reflects the number of major genes in action. In the first project, a novel and practically convenient method, called the EM-test, will be developed for testing the number of homogeneous populations in a finite mixture. Finite mixture models can be directly applied to cluster analysis, whose purpose is to segregate the data into homogeneous populations. In cluster analysis, it is important to know the number of homogeneous populations and important features for the segregation. In the second project, a criterion, which can be used to determine the number of homogenous populations and choose the important features simultaneously, will be proposed for categorical data. Switching regression models are often used to explore the relationship between a response and one or more explanatory variables when several homogeneous populations are mixed together. In the third project, testing the number of homogeneous populations by using the idea of EM-test and fitting the regression model for each homogeneous population will be explored in switching regression models. For each project, the user-friendly software will be developed. The success of this research program will lead to fundamental contributions to the methodologies and theory of finite mixture models. It will also prompt the application of finite mixture models in many fields in the natural sciences and engineering.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Alberta, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
371502-2009
Name nom
Li, Pengfei
Department département
Mathematical and Statistical Sciences
Organization id
9
Institution établissement
University of Alberta
Province en
Alberta
Province fr
Alberta
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
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
3001
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Statistical theory
Research subject fr
Théorie de statistique
Partie
2
Nb partie
5
Application title
Finite mixture models and their applications
Keyword
NSERC
Application summary
Finite mixture models are widely applied to data sets in which several homogeneous populations are mixed together. In financial economics, mixtures of three normal distributions are often applied to daily stock returns due to the existence of three risk regimes: low, normal and high. The number of homogeneous populations in a finite mixture is of great interest in applications. For example in statistical genetics, this number reflects the number of major genes in action. In the first project, a novel and practically convenient method, called the EM-test, will be developed for testing the number of homogeneous populations in a finite mixture. Finite mixture models can be directly applied to cluster analysis, whose purpose is to segregate the data into homogeneous populations. In cluster analysis, it is important to know the number of homogeneous populations and important features for the segregation. In the second project, a criterion, which can be used to determine the number of homogenous populations and choose the important features simultaneously, will be proposed for categorical data. Switching regression models are often used to explore the relationship between a response and one or more explanatory variables when several homogeneous populations are mixed together. In the third project, testing the number of homogeneous populations by using the idea of EM-test and fitting the regression model for each homogeneous population will be explored in switching regression models. For each project, the user-friendly software will be developed. The success of this research program will lead to fundamental contributions to the methodologies and theory of finite mixture models. It will also prompt the application of finite mixture models in many fields in the natural sciences and engineering.