On the effects of using the Grassmann-Taksar-Heyman method in iterative aggregation-disaggregation

被引:21
作者
Dayar, T [1 ]
Stewart, WJ [1 ]
机构
[1] N CAROLINA STATE UNIV, DEPT COMP SCI, RALEIGH, NC 27695 USA
关键词
Markov chains; decomposability; stationary probability; aggregaation-disaggregation; Gaussian elimination; sparsity schemes;
D O I
10.1137/0917021
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
Iterative aggregation-disaggregation (IAD) is an effective method for solving finite nearly completely decomposable (NCD) Markov chains. Small perturbations in the transition probabilities of these chains may lead to considerable changes in the stationary probabilities; NCD Markov chains are known to be ill-conditioned. During an IAD step, this undesirable condition is inherited by the coupling matrix arid one confronts the problem of finding the stationary probabilities of a stochastic matrix whose diagonal elements are close to 1. Tn this paper, the effects of using the Grassmann-Taksar-Heyman (GTH) method to solve the coupling matrix formed in the aggregation step are investigated. Then the idea is extended in such a way that the same direct method can be incorporated into the disaggregation step. Finally, the effects of using the GTH method in the IAD algorithm on various examples are demonstrated, and the conditions under which it should be employed are explained.
引用
收藏
页码:287 / 303
页数:17
相关论文
共 23 条
[1]  
[Anonymous], 1979, NONNEGATIVE MATRICES
[2]   ITERATIVE AGGREGATION DISAGGREGATION TECHNIQUES FOR NEARLY UNCOUPLED MARKOV-CHAINS [J].
CAO, WL ;
STEWART, WJ .
JOURNAL OF THE ACM, 1985, 32 (03) :702-719
[3]  
COURTOIS PJ, 1977, DECOMPOSABILITY
[4]   REGENERATIVE ANALYSIS AND STEADY-STATE DISTRIBUTIONS FOR MARKOV-CHAINS [J].
GRASSMANN, WK ;
TAKSAR, MI ;
HEYMAN, DP .
OPERATIONS RESEARCH, 1985, 33 (05) :1107-1116
[5]   COMPARISON OF SOME DIRECT METHODS FOR COMPUTING STATIONARY DISTRIBUTIONS OF MARKOV-CHAINS [J].
HARROD, WJ ;
PLEMMONS, RJ .
SIAM JOURNAL ON SCIENTIFIC AND STATISTICAL COMPUTING, 1984, 5 (02) :453-469
[6]   FURTHER COMPARISONS OF DIRECT METHODS FOR COMPUTING STATIONARY DISTRIBUTIONS OF MARKOV-CHAINS [J].
HEYMAN, DP .
SIAM JOURNAL ON ALGEBRAIC AND DISCRETE METHODS, 1987, 8 (02) :226-232
[7]   ITERATIVE METHODS FOR COMPUTING STATIONARY DISTRIBUTIONS OF NEARLY COMPLETELY DECOMPOSABLE MARKOV-CHAINS [J].
KOURY, JR ;
MCALLISTER, DF ;
STEWART, WJ .
SIAM JOURNAL ON ALGEBRAIC AND DISCRETE METHODS, 1984, 5 (02) :164-186
[9]   STOCHASTIC COMPLEMENTATION, UNCOUPLING MARKOV-CHAINS, AND THE THEORY OF NEARLY REDUCIBLE SYSTEMS [J].
MEYER, CD .
SIAM REVIEW, 1989, 31 (02) :240-272
[10]   ENTRYWISE PERTURBATION-THEORY AND ERROR ANALYSIS FOR MARKOV-CHAINS [J].
OCINNEIDE, CA .
NUMERISCHE MATHEMATIK, 1993, 65 (01) :109-120