NONPARAMETRIC ESTIMATION OF THE SERVICE TIME DISTRIBUTION IN THE MIG/∞ QUEUE

被引：15

作者：

Goldenshluger, Alexander ^{[1
]}

机构：

[1] Univ Haifa, Dept Stat, IL-31999 Haifa, Israel

来源：

ADVANCES IN APPLIED PROBABILITY | 2016年 / 48卷 / 04期

基金：

以色列科学基金会;

关键词：

M/G/infinity queue; nonparametric estimation; minimax risk; stationary process; covariance function; rates of convergence; M/G/INFINITY QUEUE; INFERENCE;

D O I：

10.1017/apr.2016.67

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：

The subject of this paper is the problem of estimating the service time distribution of the M/G/infinity queue from incomplete data on the queue. The goal is to estimate G from observations of the queue-length process at the points of the regular grid on a fixed time interval. We propose an estimator and analyze its accuracy over a family of target service time distributions. An upper bound on the maximal risk is derived. The problem of estimating the arrival rate is considered as well.

引用

页码：1117 / 1138

页数：22

共 26 条

[1]

[Anonymous], 1962, Stochastic processes holden-day

[2] FLUCTUATIONS OF TELEPHONE TRAFFIC [J].