进口对辽宁省经济增长贡献度的实证分析——基于时变参数卡尔曼滤波的估计

为了探寻辽宁对外经济发展的合理模式，从进口与经济增长间的关系入手，选用1985～2007年辽宁省的进口和地区生产总值数据，利用时变参数的分析方法对二者进行相关性分析，得出了进口对辽宁省的经济增长存在促进作用，而且这种作用随时递增的结论。因此，辽宁省应充分发挥进口对经济增长的积极作用，以快速实现辽宁老工业基地的振兴。     

Reml and best linear unbiased prediction in state space models

This article takes a hierarchical model approach to the estimation of state space models with diffuse initial conditions. An initial state is said to be diffuse when it cannot be assigned a proper prior distribution. In state space models this occurs either when fixed effects are present or when modelling nonstationarity in the state transition equation. Whereas much of the literature views diffuse states as an initialization problem, we follow the approach of Sallas and Harville (1981,1988) and incorporate diffuse initial conditions via noninformative prior distributions into hierarchical linear models. We apply existing results to derive the restricted loglike-lihood and appropriate modifications to the standard Kalman filter and smoother. Our approach results in a better understanding of De Jong's (1991) contributions. This article also shows how to adjust the standard Kalman filter, the fixed inter- val smoother and the state space model forecasting recursions, together with their mean square errors, for he presence of diffuse components. Using a hierarchical model approach it is shown that the estimates obtained are Best Linear Unbiased Predictors (BLUP).     

GPS/SINS组合导航系统的多传感器最优融合算法

对GPS/SINS组合导航的工作原理和模型建立进行详细的分析,设置了两个传感器。提出一种基于信息融合的导航参数最优估计滤波方法。通过计算机仿真说明该方法可提高导航系统的计算精度和速度,有较好的容错性和环境适应性,具有实际使用价值。     

Applied state space modelling of non-Gaussian time series using integration-based Kalman filtering

The main topic of the paper is on-line filtering for non-Gaussian dynamic (state space) models by approximate computation of the first two posterior moments using efficient numerical integration. Based on approximating the prior of the state vector by a normal density, we prove that the posterior moments of the state vector are related to the posterior moments of the linear predictor in a simple way. For the linear predictor Gauss-Hermite integration is carried out with automatic reparametrization based on an approximate posterior mode filter. We illustrate how further topics in applied state space modelling, such as estimating hyperparameters, computing model likelihoods and predictive residuals, are managed by integration-based Kalman-filtering. The methodology derived in the paper is applied to on-line monitoring of ecological time series and filtering for small count data.     

SPLINE SMOOTHER AS A DYNAMIC LINEAR MODEL

The paper shows how a finite dimensional representation of a cubic smoothing spline can be put in the framework of a dynamic linear model. The formulation provides an updating scheme when observations do not occur sequentially in time or space.     

饱和增长趋势预测的自适应方法

讨论了商业、市场中大量存在的饱和增长过程的自适应预测问题,详细分析了自适应增长趋势预测尤其是Kalman滤波方法涉及的模型、模型变换、递推初值确定及趋势外椎等技术问题。     

Convergence of Discount Time Series Dynamic Linear Models

This article studies the limiting behavior of multiple discount time series dynamic linear models (TSDLMs). It is shown that, under mild conditions, all discount TSDLMs converge to the constant (time-invariant) TSDLM. In particular, the limiting posterior precision matrix of the superposition of multiple discount TSDLMs is explored. For non seasonal models, the elements of the limiting posterior precision of the states are given in a recurrence relationship, while for seasonal models the solution of a linear system provides the elements of the respective limiting precision matrix. The proposed methodology uses canonical Jordan forms and it is illustrated with a detailed example of simulated data featuring both trend and seasonal time series.     

Statistical Properties of Model-Based Signal Extraction Diagnostic Tests

The problem of choosing the loss function in the Bayesian problem of many hypotheses testing is considered. It is shown that linear and quadratic loss functions are the most-used ones. For any kind of loss function, the risk function in the Bayesian problem of many hypotheses testing contains the errors of two kinds. The Bayesian decision rule minimizes the total effect of these errors. The share of each of them in the optimal value of risk function is unknown. When solving many important problems, the results caused by different errors significantly differ from each other. Therefore, it is necessary to guarantee the limitation on the most undesirable kind of these errors and to minimize the errors of the second kind. For solving these problems, this article are states and solves conditional Bayesian tasks of testing many hypotheses. The results of sensitivity analysis of the classical and conditional Bayesian problems are given and their advantages and drawbacks are considered.