Grant: Discovery Grants Program - Individual
University of Waterloo
$18,355.00
On random network inference using bootstrap
EF 2014-15
Résumé
Nowadays, with the explosive growth of online social networking and recent advances on modeling of complex and massive datasets, there exists an ever-rising interest in development of new statistical and computational methods for network analysis. Indeed, while probability models on graphs become increasingly popular to describe a variety of modern disparate social systems and natural phenomena (ranging from media propagation and social communication channels to dynamics of toxic financial assets, outbreaks of infectious diseases and spread of computer viruses), statistical methods for estimation and inference of such random network models are yet limited and substantially undeveloped. In this research program we propose an alternative nonparametric approach to network inference under limited or missing information that may be viewed as a fusion of modern resampling procedures for temporally and spatially dependent processes with classical network methodology. In particular, we suggest to adapt the ideas of nonparametric block-bootstrap and subsampling in time series analysis and propose the novel so-called ""patchwork""-bootstrap on random networks that can be employed for development of an extensive range of network topology test statistics, providing a data-driven characterization of the network structure and dynamics, and comprehensively quantifying uncertainty in random network modeling and estimation. The proposed program will not only illuminate novel data-driven perspectives for inference on random networks, which will improve accuracy of network analysis and enhance across-fertilization of ideas among modern statistical science and network theory, it will also broadly impact Canadian society by advancement of our understanding of numerous network-driven processes, from news propagation and its relation to disaster awareness and national security, to effective immunization for computer networks and enhanced marketing strategies of Canadian financial institutions. Moreover, the project will strengthen cross-disciplinary training and pave new career paths for Canadian graduates, hence increasing their global competitiveness.
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
312350-2013
Name nom
Gel, Yulia
Department département
Statistics and Actuarial Science
Organization id
33
Institution établissement
University of Waterloo
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2014
Competition year année de concours
2013
Award amount
18355
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
803
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer communications
Area of application fr
Téléinformatique
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
2
Application title
On random network inference using bootstrap
Keyword
NSERC
Application summary
Nowadays, with the explosive growth of online social networking and recent advances on modeling of complex and massive datasets, there exists an ever-rising interest in development of new statistical and computational methods for network analysis. Indeed, while probability models on graphs become increasingly popular to describe a variety of modern disparate social systems and natural phenomena (ranging from media propagation and social communication channels to dynamics of toxic financial assets, outbreaks of infectious diseases and spread of computer viruses), statistical methods for estimation and inference of such random network models are yet limited and substantially undeveloped. In this research program we propose an alternative nonparametric approach to network inference under limited or missing information that may be viewed as a fusion of modern resampling procedures for temporally and spatially dependent processes with classical network methodology. In particular, we suggest to adapt the ideas of nonparametric block-bootstrap and subsampling in time series analysis and propose the novel so-called ""patchwork""-bootstrap on random networks that can be employed for development of an extensive range of network topology test statistics, providing a data-driven characterization of the network structure and dynamics, and comprehensively quantifying uncertainty in random network modeling and estimation. The proposed program will not only illuminate novel data-driven perspectives for inference on random networks, which will improve accuracy of network analysis and enhance across-fertilization of ideas among modern statistical science and network theory, it will also broadly impact Canadian society by advancement of our understanding of numerous network-driven processes, from news propagation and its relation to disaster awareness and national security, to effective immunization for computer networks and enhanced marketing strategies of Canadian financial institutions. Moreover, the project will strengthen cross-disciplinary training and pave new career paths for Canadian graduates, hence increasing their global competitiveness.