COUNTING PROCESSES WITH BERNSTEIN INTERTIMES AND RANDOM JUMPS
被引:25
|
作者:
Orsingher, Enzo
论文数: 0引用数: 0
h-index: 0
机构:
Univ Roma La Sapienza, Dipartimento Stat, Piazzale Aldo Moro 5, I-00185 Rome, ItalyUniv Roma La Sapienza, Dipartimento Stat, Piazzale Aldo Moro 5, I-00185 Rome, Italy
Orsingher, Enzo
[1
]
Toaldo, Bruno
论文数: 0引用数: 0
h-index: 0
机构:
Univ Roma La Sapienza, Dipartimento Stat, Piazzale Aldo Moro 5, I-00185 Rome, ItalyUniv Roma La Sapienza, Dipartimento Stat, Piazzale Aldo Moro 5, I-00185 Rome, Italy
Toaldo, Bruno
[1
]
机构:
[1] Univ Roma La Sapienza, Dipartimento Stat, Piazzale Aldo Moro 5, I-00185 Rome, Italy
Levy measure;
Bernstein function;
subordinator;
negative binomial;
beta random variable;
FRACTIONAL POISSON PROCESSES;
D O I:
10.1017/S0021900200113063
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:
In this paper we consider point processes N-f (t), t > 0, with independent increments and integer-valued jumps whose distribution is expressed in terms of Bernstein functions f with Levy measure nu. We obtain the general expression of the probability generating functions G(f) of N-f, the equations governing the state probabilities p(k)(f) of N-f, and their corresponding explicit forms. We also give the distribution of the first-passage times T-k(f) of N-f, and the related governing equation. We study in detail the cases of the fractional Poisson process, the relativistic Poisson process, and the gamma-Poisson process whole state probabilities have the form of a negative binomial. The distribution of the times tau(lj)(j) of jumps with height l(j) (Sigma(r)(j=1) l(j) = k) under the condition N(t) = k for all these special processes is investigated in detail.