Structure selection for stochastic processes in high dimensions

• Grant number: 16/17394-0
• Support Opportunities: Regular Research Grants
• Duration: March 01, 2017 - May 31, 2019
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Florencia Graciela Leonardi
• Grantee: Florencia Graciela Leonardi
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated researchers: Alexsandro Giacomo Grimbert Gallo ; Miguel Natalio Abadi ; Roberto Imbuzeiro Moraes Felinto de Oliveira

Abstract

The main goal of this research project is to study methods of statistical inference to identify different structural aspects in stochastic processes in high dimensions. We will consider principally the estimation of the graph of interaction in Markov random fields, also known in the statistical literature as graphical models, and the identification of change points for stochastic processes, both discrete and continuous. These two lines of research, particularly for models in high dimensions, constitute natural extensions of recent results obtained by me and my co-authors in the sabbatical research period in the Swiss Institute of Technology at Zurich. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

CASTRO, BRUNO M.; LEMES, RENAN B.; CESAR, JONATAS; HUNEMEIER, TABITA; LEONARDI, FLORENCIA. A model selection approach for multiple sequence segmentation and dimensionality reduction. JOURNAL OF MULTIVARIATE ANALYSIS, v. 167, p. 319-330, SEP 2018. (13/07699-0, 16/17394-0)