Hypothesis tests in fixed and mixed effects models and their applications on predictors in ANCOVA and nonparametric regression

• Grant number: 12/22603-6
• Support type: Research Grants - Visiting Researcher Grant - International
• Duration: February 25, 2013 - March 18, 2013
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal researcher: Ronaldo Dias
• Grantee: Ronaldo Dias
• Visiting researcher: Michael G. Akritas
• Visiting researcher institution: Pennsylvania State University, United States
• Home Institution: Instituto de Matemática, Estatística e Computação Científica (IMECC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

Let X be a vector of available covariates and Y be the response variable. Under the nonparametric model Y = m(X)+(X), Zambom and Akritas (2012) proposed a test based on methods of one-way ANOVA when the number of factor levels goes to innity. Moreover, using the results of this test Zambom e Akritas suggested a variable selection algorithm using FDR (False Discovery Control). The algorithm has good results in simulations, but its asymptotic properties are yet to be explored. Therefore, the main objective of this project is to study the theoretical asymptotic consistency of this method. Other objectives include the study of other methodologies that allow this type of hypothesis test and variable selection in mixed models. (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)

ZAMBOM, ADRIANO ZANIN; AKRITAS, MICHAEL G. NONPARAMETRIC LACK-OF-FIT TESTING AND CONSISTENT VARIABLE SELECTION. STATISTICA SINICA, v. 24, n. 4, p. 1837-1858, OCT 2014. Web of Science Citations: 9.