New regression models to data set with binary and/or bounded response

• Grant number: 17/15452-5
• Support Opportunities: Scholarships abroad - Research
• Start date: March 01, 2018
• End date: February 28, 2019
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Jorge Luis Bazan Guzman
• Grantee: Jorge Luis Bazan Guzman
• Host Investigator: Dipak Kumar Dey
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: University of Connecticut (UCONN), United States

Abstract

The main objective of this project is to propose different regression models for binary and/orlimited responses in the unit interval studying aspects of inference and modeling when consideredreal data. Specifically, the project aims to develop new regression models to the case of discrete limited responses in the unit interval considering by example proposals of new links for binary regression models. Besides that, we studied the case of new regression models to continuouslimited response. Extensions this models to ordinal regression model, mixed regression modeland item response theory are also considered. . Simulation and application studies to real data complete the objectives of this project. The proposal is justified by the shortage of research that accommodates such kind of data, the practical implications of the results of such modeling, by the relevance of the working together with Prof. Dipak Dey and to reinforce the contact with the Department of Statistics at the University Of Connecticut. It is expected publish paper in relevant international journals, made presentations in scientific meetings in order to disseminate the results obtained as well as the institutions involved. Likewise, it is hoped the reinforce of the partnership between the two universities, with the aim of establishing new projects including the participation of graduate students.

Scientific publications

(8)

(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)

DA SILVA, MARCELO A.; HUGGINS-MANLEY, ANNE C.; MAZZON, JOSE A.; BAZAN, JORGE L.. Bayesian estimation of a flexible bifactor generalized partial credit model to survey data. Journal of Applied Statistics, v. 46, n. 13, p. 1-16, MAR 2019. (17/15452-5)

DA SILVA, MARCELO A.; LIU, REN; HUGGINS-MANLEY, ANNE CORINNE; BAZAN, JORGE L.. Bayesian estimation of multidimensional polytomous item response theory models with Q-matrices using Stan. COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, SEP 2021. (17/15452-5)

DA SILVA, MARCELO A.; LIU, REN; HUGGINS-MANLEY, ANNE C.; BAZAN, JORGE L.. Incorporating the Q-Matrix Into Multidimensional Item Response Theory Models. EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT, v. 79, n. 4, p. 665-687, AUG 2019. (17/15452-5)

FLORES, SANDRA E.; PRATES, MARCOS O.; BAZAN, JORGE L.; BOLFARINE, HELENO B.. Spatial regression models for bounded response variables with evaluation of the degree of dependence. STATISTICS AND ITS INTERFACE, v. 14, n. 2, p. 95-107, 2021. (17/15452-5)

BAZAN, JORGE LUIS. Four comments for the paper from Liu et al. (2019). JOURNAL OF THE KOREAN STATISTICAL SOCIETY, v. 48, n. 4, p. 2-pg., 2019-12-01. (17/15452-5)

DA PAZ, ROSINEIDE F.; BALAKRISHNAN, NARAYANASWAMY; BAZAN, JORGE LUIS. L-Logistic regression models: Prior sensitivity analysis, robustness to outliers and applications. BRAZILIAN JOURNAL OF PROBABILITY AND STATISTICS, v. 33, n. 3, p. 455-479, AUG 2019. (17/15452-5)

HUAYANAY, ALEX DE LA CRUZ; BAZAN, JORGE L.; CANCHO, VICENTE G.; DEY, DIPAK K.. Performance of asymmetric links and correction methods for imbalanced data in binary regression. JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, v. 89, n. 9, p. 1694-1714, JUN 13 2019. (17/15452-5)

DA PAZ, ROSINEIDE; BAZAN, JORGE LUIS; LACHOS, VICTOR HUGO; DEY, DIPAK. A finite mixture mixed proportion regression model for classification problems in longitudinal voting data. Journal of Applied Statistics, NOV 2021. (17/15452-5)