Point and interval estimation in the hierarchic logistic model two-level

Artur José LEMONTE[1]

Tatiane Ferreira Nascimento Melo da SILVA[2]

§    ABSTRACT: A study on parameter estimation in the two-level hierarchical logistic model is presented in this paper. Estimators based on first-order marginal quasi-likelihood were taken into account, and their estimation algorithm can be found in Goldstein (1991). The performance of these estimators was evaluated through exact and interval estimation. Additionally, the cover rates of the confidence intervals and power of the significance tests are also presented. Monte Carlo simulation was used, and finally, an application of real data is presented and discussed.

§    KEYWORDS: Hierarchic models; interval estimation; logistc model; Monte Carlo simulation; punctual estimation; random effect.

[1]Departamento de Estatística, Universidade Federal de Pernambuco - UFPE, CEP 50740-540, Recife, PE, Brasil. E-mail: artur@cox.de.ufpe.br

[2]Instituto de Matemática e Estatística, Universidade de São Paulo - USP, CEP 05508-090, São Paulo, SP, Brasil. E-mail: tafename@ime.usp.br