Statistical Estimation for a Reliability Model Based on Shot-Noise Processes in a Case of Small Samples

Abstract

A failure model based on shot-noise processes is considered. A discussed reliability system is subjected to shocks. Shocks of random magnitude occur according to a homogeneous Poisson process and decay with time according to the deterministic function h(t). The rate of the Poisson process and the distribution of magnitude values are unknown. Two sample populations are available: a sample of intervals between shocks and a sample of magnitude values. The case of small samples is considered. The purpose is to estimate the expectation of the system stress level at time t. We consider the plug-in and resampling estimators of the above mentioned characteristics. The expectation and variance of the suggested estimators are investigated. The numerical examples and simulation studies show that the resampling estimator has some advantages.

Key Words:

• Estimation
• expectation and variance
• plug-in estimator
• resampling
• shot noise process

Additional information

Notes on contributors

Jelena Fiosina

Jelena Fiosina is a researcher in Clausthal Technical University, Germany. She obtained her Ph.D. from Transport and Telecommunication Institute, Latvia. Her research interests include computational statistics, data analysis, stochastic processes and modeling. She has a number of publications in reviewed conference proceedings and journals, including Journal of Statistical Planning and Inference, Advances in Intelligent and Soft Computing.

Maksims Fiosins

Maksims Fiosins is a researcher in Clausthal Technical University, Germany. He obtained his Ph.D. from Riga Technical University, Latvia. His research interests include computational statistics, data analysis, decision making and modeling. He has a number of publications in reviewed conference proceedings and journals, including Journal of Statistical Planning and Inference, Advances in Intelligent and Soft Computing.