Grant: Discovery Grants Program - Individual
University of Ottawa
$10,000.00
Martingale and multivariate asymptotic results for analysis of longtudinal data
FY 2009-10
Summary
This research program uses techniques from various areas of statistics and probability, combined in an innovative way to offer novel formulations with solutions, or to obtain better solutions to specified practical problems.
(a) Generalized estimating equations(GEEs) provide a semi-parametric approach to longitudinal data analysis. We study the strong consistency, asymptotic normality and aymptotic first-order efficiency of a sequence of GEE estimators. A class of martingale transforms is required and so the full power of martingale theory becomes available, which would result in simpler conditions and proofs. Random covariates will also be considered, which is of interest in control theory, and missing data, of interest in data analysis.
(b) We address the simultaneous and sequential selection of samples for designs commonly use in business surveys. Maximization of the expected overlap of samples reduces the costs of surveys and the variability of estimators over time; minimization decreases response burden. We will study the complex case of strata jumpers and the difficult combinatorial problem of minimizing the overlap for several surveys.
(c) A product probability space was recently defined, which accomodates design and model-based inference. The porposed project would complete this definition to include Bayesian inference. A complete description of conditioning in this product space will also be given and one would be able to provide precise definitions of concepts like informative or noninformative sampling.
Program
Discovery Grants Program - Individual
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
185679-2008
Name nom
SchiopuKratina, Ioana
Department département
Mathematics and Statistics
Organization id
28
Institution établissement
University of Ottawa
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2009
Competition year année de concours
2008
Award amount
10000
Program id
RGPIN
Program nane 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
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
Martingale and multivariate asymptotic results for analysis of longtudinal data
Keyword
NA
Application summary
This research program uses techniques from various areas of statistics and probability, combined in an innovative way to offer novel formulations with solutions, or to obtain better solutions to specified practical problems.
(a) Generalized estimating equations(GEEs) provide a semi-parametric approach to longitudinal data analysis. We study the strong consistency, asymptotic normality and aymptotic first-order efficiency of a sequence of GEE estimators. A class of martingale transforms is required and so the full power of martingale theory becomes available, which would result in simpler conditions and proofs. Random covariates will also be considered, which is of interest in control theory, and missing data, of interest in data analysis.
(b) We address the simultaneous and sequential selection of samples for designs commonly use in business surveys. Maximization of the expected overlap of samples reduces the costs of surveys and the variability of estimators over time; minimization decreases response burden. We will study the complex case of strata jumpers and the difficult combinatorial problem of minimizing the overlap for several surveys.
(c) A product probability space was recently defined, which accomodates design and model-based inference. The porposed project would complete this definition to include Bayesian inference. A complete description of conditioning in this product space will also be given and one would be able to provide precise definitions of concepts like informative or noninformative sampling.