Cross-validation for selection of zero-modified Poisson models

• Grant number: 22/01972-5
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2022
• End date: January 31, 2023
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Katiane Silva Conceição
• Grantee: Brenda da Silva Muniz
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

In this project, the cross-validation technique will be considered to select the explicative variables to be included in the regression model. Assuming that the variable of interest is a counting process, the Zero-Modified Poisson distribution, which is a generalization of the Poisson distribution, will be considered for the modeling approach, since in many practical situations there may be a frequency discrepancy (high or low) of zero in the observed dataset. In this sense, we want to explain the behavior of the zero-modified count dataset, using explicative variables selected via the k-fold cross-validation method, and obtain the most appropriate regression model to describe and predict the experimental data. A simulation study will be carried out to compare the performance of the models with the considered technique, as well as to present some applications with real data sets.(AU)