Monte Carlo Approximations for General State-Space Models
针对非线性非高斯状态空间模型，提出两种基于简单拒绝算法的迭代抽样方法，用于从滤波密度和平滑密度中生成样本，并通过实例与其他方法比较。
Non-linear and non-Gaussian state space models form a very large and exible model class in time series analysis.Two methods for generating iteratively samples from lter densities and smoother densities by simple rejection algorithms are introduced.We illustrate the behavior of our methods in several non-linear and non-Gaussian examples and compare them with other wellknown methods.