This paper is devoted to a study of geometric properties of AR(q) nonlinear regression models. We present geometric frameworks for regression parameter space and autoregression parameter space respectively based on th...This paper is devoted to a study of geometric properties of AR(q) nonlinear regression models. We present geometric frameworks for regression parameter space and autoregression parameter space respectively based on the weighted inner product by fisher information matrix. Several geometric properties related to statistical curvatures are given for the models. The results of this paper extended the work of Bates & Watts(1980,1988)[1.2] and Seber & Wild (1989)[3].展开更多
This paper presents an interval effective independence method for optimal sensor placement, which contains uncertain structural information. To overcome the lack of insufficient statistic description of uncertain para...This paper presents an interval effective independence method for optimal sensor placement, which contains uncertain structural information. To overcome the lack of insufficient statistic description of uncertain parameters, this paper treats uncertainties as non-probability intervals. Based on the iterative process of classical effective independence method, the proposed study considers the eliminating steps with uncertain cases. Therefore, this method with Fisher information matrix is extended to interval numbers, which could conform to actual engineering. As long as we know the bounds of uncertainties, the interval Fisher information matrix could be obtained conveniently by interval analysis technology. Moreover, due to the definition and calculation of the interval relationship, the possibilities of eliminating candidate sensors in each iterative process and the final layout of sensor placement are both presented in this paper. Finally, two numerical examples, including a five-storey shear structure and a truss structure are proposed respectively in this paper. Compared with Monte Carlo simulation, both of them can indicate the veracity of the interval effective independence method.展开更多
基金Supported by the NSSFC(02BTJ001) Supported by the NSSFC(04BTJ002) Supported by the Grant for Post-Doctorial Fellows in Southeast University
文摘This paper is devoted to a study of geometric properties of AR(q) nonlinear regression models. We present geometric frameworks for regression parameter space and autoregression parameter space respectively based on the weighted inner product by fisher information matrix. Several geometric properties related to statistical curvatures are given for the models. The results of this paper extended the work of Bates & Watts(1980,1988)[1.2] and Seber & Wild (1989)[3].
基金supported by the National Natural Science Foundation of China(Grant No.11502278)
文摘This paper presents an interval effective independence method for optimal sensor placement, which contains uncertain structural information. To overcome the lack of insufficient statistic description of uncertain parameters, this paper treats uncertainties as non-probability intervals. Based on the iterative process of classical effective independence method, the proposed study considers the eliminating steps with uncertain cases. Therefore, this method with Fisher information matrix is extended to interval numbers, which could conform to actual engineering. As long as we know the bounds of uncertainties, the interval Fisher information matrix could be obtained conveniently by interval analysis technology. Moreover, due to the definition and calculation of the interval relationship, the possibilities of eliminating candidate sensors in each iterative process and the final layout of sensor placement are both presented in this paper. Finally, two numerical examples, including a five-storey shear structure and a truss structure are proposed respectively in this paper. Compared with Monte Carlo simulation, both of them can indicate the veracity of the interval effective independence method.
