ON RATES OF CONVERGENCE OF STOCHASTIC RELAXATION FOR GAUSSIAN AND NON-GAUSSIAN DISTRIBUTIONS

被引:41
作者
AMIT, Y
机构
关键词
MONTE-CARLO; STOCHASTIC RELAXATION; CONVERGENCE RATES; ANGLES BETWEEN SUBSPACES; WEINER-ITO DECOMPOSITION;
D O I
10.1016/0047-259X(91)90033-X
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
We obtain rates of convergence of stochastic relaxation (heat bath algorithm) for continuous densities which have the form of bounded perturbations of Gaussian densities. The rates are calculated in the spaces of square integrable functions with respect to these desities in which the operator generated by the stochastic relaxation process has the form of a product of projections. © 1991.
引用
收藏
页码:82 / 99
页数:18
相关论文
共 12 条