In this paper, a meshless method is introduced for NDT computation. Compared with the conventional FEM, it can avoid the onerous mesh generation and updating, only a distribution of points and the description of the b...In this paper, a meshless method is introduced for NDT computation. Compared with the conventional FEM, it can avoid the onerous mesh generation and updating, only a distribution of points and the description of the boundaries are needed. The mathematical background for moving least square approximation employed in the method is given, and the numerical implementation is discussed. Application of the method for MFL computation and comparison with the results from FEM are also presented.展开更多
The purpose of this paper is to test the underlying serial correlation in a partially linear single-index model. Under mild conditions, the proposed test statistics are shown to have standard chi- squared distribution...The purpose of this paper is to test the underlying serial correlation in a partially linear single-index model. Under mild conditions, the proposed test statistics are shown to have standard chi- squared distribution asymptotically when there is no serial correlation in the error terms. To illustrate their finite sample properties, simulation experiments, as well as a real data example, are also provided. It is revealed that the finite sample performances of the proposed test statistics are satisfactory in terms of both estimated sizes and powers.展开更多
文摘In this paper, a meshless method is introduced for NDT computation. Compared with the conventional FEM, it can avoid the onerous mesh generation and updating, only a distribution of points and the description of the boundaries are needed. The mathematical background for moving least square approximation employed in the method is given, and the numerical implementation is discussed. Application of the method for MFL computation and comparison with the results from FEM are also presented.
基金supported by CCNU under Grant No.09A01002the SCR of Chongqing Municipal Education Commission under Grant No.KJ110713the National Natural Science Foundation of China under Grant Nos.11101452 and 71172093
文摘The purpose of this paper is to test the underlying serial correlation in a partially linear single-index model. Under mild conditions, the proposed test statistics are shown to have standard chi- squared distribution asymptotically when there is no serial correlation in the error terms. To illustrate their finite sample properties, simulation experiments, as well as a real data example, are also provided. It is revealed that the finite sample performances of the proposed test statistics are satisfactory in terms of both estimated sizes and powers.
