摘要:In this paper, we review the score test procedure used for testing polynomial covariatee.ects in a semi parametric additive mixed model. This test is based on the mixedmodel representation of the smoothing spline estimator of the nonparametric functionand treating the inverse of the smoothing parameter as an extra variance component.Zhang and Lin (2003) found that the score test of polynomial test for non Gaussianresponses follows a scaled chi-squared distribution. Simulation studies showed thattheir approximation is less satisfactory for binary data. To overcome this deficiency,we apply the technique of Monte Carlo in order to obtain provably exact pro cedures.Derivation and performance of each testing procedure are discussed throughout thesimulations that we conducted
关键词:semi parametric additive mixed models; polynomial test; score test; Monte Carlo test