Consistency of the Bayesian Information Criterion for Hidden Markov Models

• Grant number: 24/13253-9
• Support Opportunities: Scholarships in Brazil - Doctorate (Direct)
• Start date: September 01, 2024
• End date: August 31, 2028
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics
• Principal Investigator: Florencia Graciela Leonardi
• Grantee: Marília de Melo Sombra
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 23/13453-5 - Stochastic systems modeling, AP.TEM

Abstract

The main goal of this project is to study the problem of order estimation in the class of Hidden Markov models. This is a classic problem in the inference literature for stochastic processes, with some advances towards order estimation. However, the consistency of the Bayesian Information Criterion (BIC) still remains an open question. This study, which will be conducted based on recent results from the research group obtained for other models, such as the stochastic block model, is relevant in the area and could lead to original results with great impact.