Grant: Discovery Grants Program - Individual
University of Windsor
$28,000.00
Analysis of discrete/Survival data, longitudinal or otherwise, with over-dispersion, zero-inflation, missing response, missing covariate information, covariate measurement errors and bias correction
EF 2024-25
Résumé
Analysis of discrete Data (counts/Proportions/regression models), survival Data, and longitudinal data with over dispersion, zero inflation, missing responses, missing covariate information, covariate measurement error, and Bias Correction
Summary of Proposal
I am planning to deal with a related group of problems: Analysis of discrete/Survival data, longitudinal or otherwise, with over dispersion, zero inflation, missing response, missing covariate information, covariate measurement errors, and bias correction.
Discrete data in the form of counts/ data in the form of proportions/survival data often exhibit extra variation that cannot be explained by a simple model, such as the binomial or the Poisson model or an exponential model. In the case of discrete data or data in the form of proportions (univariate or multivariate) these data, sometimes, show more zero counts than what can be predicted by a simple model. Therefore, a discrete generalized linear model (Poisson or binomial) may fail to fit a set of discrete data either because of zero inflation or because of over dispersion or because there is zero inflation as well as over dispersion in the data. My work in this area and those of my HPQs involves finding suitable model/models through testing of goodness fit (Paul and Deng, R29, R30; Mian and Paul, C12; Alam and Paul, C11; Ozonur and Paul, C5; and Pal, Luo and Bagui, and Paul (C4). A lot of work in this area remains to be done. Bias correction of the MLEs, specifically, of the over dispersion and zero inflation parameters are important as the performance of the test statistics improves when bias corrected estimates are used (Zhang, Paul and Wang, C6). Many problems involving bias correction in MLE or MLE type estimates (such as quasi likelihood, extended quasi likelihood, double extended quasi likelihood estimates) need to be investigated. Further, estimation of parameters and removing bias of the estimates when regression models of discrete data or survival data, data in the form of proportions, count data in the form ANOVA Tables (one way/two way, balanced/unbalanced) involve missing responses as well as missing covariate information and covariate measurement errors need investigation. We have made some progress recently in estimation and testing in discrete data models and models for data in the form of proportions in presence of missing response data or data involving missing covariate information (Mian and Paul, R27, Luo and Paul, C10, Mian and Paul, C3). Longitudinal data arise in Health Studies, Epidemiological studies, and other similar areas of applications. For efficient estimation and joint modelling of means, variances, and correlations in Longitudinal Data Regression Analysis using semi-parametric modelling is important. Recently Mamun and Paul (C1, C2) made some progress. Another area is developing inference procedures in multivariate discrete model with over-dispersion zero-inflation. All these will be developed.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
RGPIN-2024-03805
Name nom
Paul, Sudhir
Department département
mathematics and statistics
Organization id
37
Institution établissement
University of Windsor
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
28000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et statistique
Area of application code
1211
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Medical and health sciences
Area of application fr
Sciences médicales et de la santé
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
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Analysis of discrete/Survival data, longitudinal or otherwise, with over-dispersion, zero-inflation, missing response, missing covariate information, covariate measurement errors and bias correction
Keywords
discrete data (count/proportion); survival data; longitudinal data; over-dispersion; zero-inflation; missing data; covariate information; covariate measurement error; bias correction; regression model
Application summary
Analysis of discrete Data (counts/Proportions/regression models), survival Data, and longitudinal data with over dispersion, zero inflation, missing responses, missing covariate information, covariate measurement error, and Bias Correction
Summary of Proposal
I am planning to deal with a related group of problems: Analysis of discrete/Survival data, longitudinal or otherwise, with over dispersion, zero inflation, missing response, missing covariate information, covariate measurement errors, and bias correction.
Discrete data in the form of counts/ data in the form of proportions/survival data often exhibit extra variation that cannot be explained by a simple model, such as the binomial or the Poisson model or an exponential model. In the case of discrete data or data in the form of proportions (univariate or multivariate) these data, sometimes, show more zero counts than what can be predicted by a simple model. Therefore, a discrete generalized linear model (Poisson or binomial) may fail to fit a set of discrete data either because of zero inflation or because of over dispersion or because there is zero inflation as well as over dispersion in the data. My work in this area and those of my HPQs involves finding suitable model/models through testing of goodness fit (Paul and Deng, R29, R30; Mian and Paul, C12; Alam and Paul, C11; Ozonur and Paul, C5; and Pal, Luo and Bagui, and Paul (C4). A lot of work in this area remains to be done. Bias correction of the MLEs, specifically, of the over dispersion and zero inflation parameters are important as the performance of the test statistics improves when bias corrected estimates are used (Zhang, Paul and Wang, C6). Many problems involving bias correction in MLE or MLE type estimates (such as quasi likelihood, extended quasi likelihood, double extended quasi likelihood estimates) need to be investigated. Further, estimation of parameters and removing bias of the estimates when regression models of discrete data or survival data, data in the form of proportions, count data in the form ANOVA Tables (one way/two way, balanced/unbalanced) involve missing responses as well as missing covariate information and covariate measurement errors need investigation. We have made some progress recently in estimation and testing in discrete data models and models for data in the form of proportions in presence of missing response data or data involving missing covariate information (Mian and Paul, R27, Luo and Paul, C10, Mian and Paul, C3). Longitudinal data arise in Health Studies, Epidemiological studies, and other similar areas of applications. For efficient estimation and joint modelling of means, variances, and correlations in Longitudinal Data Regression Analysis using semi-parametric modelling is important. Recently Mamun and Paul (C1, C2) made some progress. Another area is developing inference procedures in multivariate discrete model with over-dispersion zero-inflation. All these will be developed.