【摘 要】
:
In this paper, we propose a bias-corrected empirical likelihood (BCEL) for the divergiug number of parameters in a varying coefficient partially linear mode
【机 构】
:
HongKongBaptistUniversity
【出 处】
:
IMS-China International Conference on Statistics and Probabi
论文部分内容阅读
In this paper, we propose a bias-corrected empirical likelihood (BCEL) for the divergiug number of parameters in a varying coefficient partially linear model.Even when plug-in estimation for nonparametric coefficient functions are used, the BCEL ratio is still asymptotically normal.Thus, it can be dircctly used, without using any other point estimation, to construct confidence region for the parameters of interest.The regularity conditions on the model are very mild: other than the nsual regularity conditions on the smoothness and nonparametric estimation of involved nouparametric functions, only moment conditions on the covariates and error are required.Also the quasi-likelihood framework, least favorable curve, higher order kernel and under smoothing are avoided.Furthermore, the diverging rate of the number of the parameters can be faster than those in the literature.A simulation study is carried out to assess the performance of the proposed method and to compare it with the profile least squares method.A real dataset is analysed for illustration.
其他文献
We consider a Gelfand triple E (C) H (C) E* generated by a positive operator A and give a general form of characteristic exponents of infinitely divisible d
In clinical and epidemiological studies research are very much interested to know the inter ?observer variation in a continuous variable or two measurement
In this paper, we discuss tests of heteroscedasticity and/or autocorrelation in nonlinear models with AR(1)and symmetrical errors.The symmetrical errors dis
The diagnostic capability or accuracy of a medical test is often assessed using a receiver operating characteristic (ROC) curve.ROC regression methodology o
We introducethe m-dependent approximation for a class of stationary processes.As its applications, under quite easy verifiable and more weaker conditions, w
Thc support vector machine (SVM) has been widely applied for classification problems in both machiue learning and statistics.Despite its popularity, however
Television audience research can provide much useful information to broadcast ers and advertisers to better understand the audience viewing behavior and imp
Large-scale simultaneous hypothesis testing problems are ubiquitous in many sci entific fields including genomic and image analysis.Many methods have been d
We develop a consistent and highly efficient marginal model for missing at ran dom data using an estimating function approach.Our approach differs from inve
We study the smooth quantile estimator based on a kernel k and a sequence of bandwidth for stationary strong mixing random variables.Under minimal assump ti