We investigate perturbation for continuous-time Markov chains(CTMCs) on a countable state space. Explicit bounds on ?D and D are derived in terms of a drift condition, where ? and D represent the perturbation of the i...We investigate perturbation for continuous-time Markov chains(CTMCs) on a countable state space. Explicit bounds on ?D and D are derived in terms of a drift condition, where ? and D represent the perturbation of the intensity matrices and the deviation matrix, respectively. Moreover, we obtain perturbation bounds on the stationary distributions, which extends the results by Liu(2012) for uniformly bounded CTMCs to general(possibly unbounded) CTMCs. Our arguments are mainly based on the technique of augmented truncations.展开更多
Let {Xn; n ∈ N2} be a two dimensionally indexed linear stationary random field generated by a 1/4 martingale difference white noise. The logarithm uniform convergency resulte for the weighted periodogram of is proved.
基金supported in part by National Natural Science Foundation of China(11371007,61370002,61403132)Multi-Year Research of University of Macao under Grants No.MYRG205(Y1-L4)-FST11-TYY and No.MYRG187(Y1-L3))-FST11-TYY+1 种基金Start-up Research of University of Macao under Grant No.SRG010-FST11-TYYNatural Science Foundation of Hubei Province(2011CDA003)
基金supported by National Natural Science Foundation of China(Grant No.11211120144)the Fundamental Research Funds for the Central Universities(Grant No.2010QYZD001)
文摘We investigate perturbation for continuous-time Markov chains(CTMCs) on a countable state space. Explicit bounds on ?D and D are derived in terms of a drift condition, where ? and D represent the perturbation of the intensity matrices and the deviation matrix, respectively. Moreover, we obtain perturbation bounds on the stationary distributions, which extends the results by Liu(2012) for uniformly bounded CTMCs to general(possibly unbounded) CTMCs. Our arguments are mainly based on the technique of augmented truncations.
文摘Let {Xn; n ∈ N2} be a two dimensionally indexed linear stationary random field generated by a 1/4 martingale difference white noise. The logarithm uniform convergency resulte for the weighted periodogram of is proved.
