Um teste de especificaÃÃo correta em modelos de regressÃo beta

AUTOR(ES)
DATA DE PUBLICAÇÃO

2007

RESUMO

The beta regression model introduced by Ferrari &Cribari-Neto (2004) is useful for modelling responses that assume values in the standard unit interval (0, 1). The beta parametrization used is based on mean and precision parameters, and the mean response is related to a linear predictor through a link function. The chief goal of this thesis is to propose a misspecification test for such a class of models. Our test is based on that proposed by Ramsey (1969) for the linear regression model. The numerical results favor the use of the score test in the implementation of our test, and the use of powers of the estimated linear predictor and of the fitted mean response as auxiliary variables

ASSUNTO(S)

simulaÃÃo de monte carlo regressÃo beta monte carlo simulation hypotheses tests reset testes de hipÃteses estatistica beta regression reset

Documentos Relacionados