Grant: Discovery Grants Program - Individual
University of Manitoba
$17,000.00
Decision theoretic inference in problems involving balanced loss functions, constraint parameter spaces and finite mixture models
EF 2013-14
Résumé
In this proposed research I will focus on decision theoretic studies related to (i) the use of balanced loss functions in a variety of contexts, (ii) estimation in constrained parameter spaces, and (iii) estimation in finite mixture models. In decision theory, the usual loss functions, such as the squared error loss, measure the precision of estimation. However, other losses representing goodness of fit may also be of interest. In the past, only one of these two types of losses was typically used in a given problem. Recently, however, balanced loss functions were introduced. These balanced loss functions are made by combining the two loss functions and balancing the two criteria according to some chosen weights. Balanced loss functions can be used as a tool for making better decisions. Exploring decision theoretic issues, new potential applications and some practical implementations of balanced loss functions are important and will be the primary focus of this research. Another important subject in statistics is statistical inference in constrained parameter spaces. Many such problems arise in practice. It often happens that the usual estimators, which have good properties in unconstrained parameter spaces, lose some of their optimal properties when the parameter spaces are constrained. Finding estimators which dominate the usual estimators such as the maximum likelihood, minimum risk equivariant or UMVUE estimators is of interest. Issues of admissibility and minimaxity also arise naturally in constrained parameter spaces. In this research proposal, I propose to study decision theoretic estimation of quantiles in constrained parameter spaces for different models and under different loss functions. This is an important problem in many area of applications such as life testing and reliability. Finally, I will continue to pursue this decision theoretic approach to some areas of applications. These include, issues in credibility premium estimation, sample surveys and modeling radar images in both Bayesian and non-Bayesian contexts.
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
386575-2010
Name nom
JafariJozani, Mohammad
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
2010
Award amount
17000
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
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
4
Nb partie
5
Application title
Decision theoretic inference in problems involving balanced loss functions, constraint parameter spaces and finite mixture models
Keyword
NSERC
Application summary
In this proposed research I will focus on decision theoretic studies related to (i) the use of balanced loss functions in a variety of contexts, (ii) estimation in constrained parameter spaces, and (iii) estimation in finite mixture models. In decision theory, the usual loss functions, such as the squared error loss, measure the precision of estimation. However, other losses representing goodness of fit may also be of interest. In the past, only one of these two types of losses was typically used in a given problem. Recently, however, balanced loss functions were introduced. These balanced loss functions are made by combining the two loss functions and balancing the two criteria according to some chosen weights. Balanced loss functions can be used as a tool for making better decisions. Exploring decision theoretic issues, new potential applications and some practical implementations of balanced loss functions are important and will be the primary focus of this research. Another important subject in statistics is statistical inference in constrained parameter spaces. Many such problems arise in practice. It often happens that the usual estimators, which have good properties in unconstrained parameter spaces, lose some of their optimal properties when the parameter spaces are constrained. Finding estimators which dominate the usual estimators such as the maximum likelihood, minimum risk equivariant or UMVUE estimators is of interest. Issues of admissibility and minimaxity also arise naturally in constrained parameter spaces. In this research proposal, I propose to study decision theoretic estimation of quantiles in constrained parameter spaces for different models and under different loss functions. This is an important problem in many area of applications such as life testing and reliability. Finally, I will continue to pursue this decision theoretic approach to some areas of applications. These include, issues in credibility premium estimation, sample surveys and modeling radar images in both Bayesian and non-Bayesian contexts.