BAYES ESTIMATORS OF PARAMETERS OF BINOMIAL TYPE RAYLEIGH CLASS SOFTWARE RELIABILITY GROWTH MODEL USING NON-INFORMATIVE PRIORS

Rajesh  Singh; Pritee  Singh; Kailash R.  Kale

Rajesh Singh Department of Statistics, S.G.B. Amravati University, Amravati, India
Pritee Singh Department of Statistics, Institute of Science, Nagpur, India
Kailash R. Kale Department of Statistics, R. D. N. College, Mumbai, India

Keywords: Binomial Process, Non-Informative Prior, Maximum Likelihood Estimator (MLE), Rayleigh Class, Software Reliability Growth Model (SRGM)

Abstract

In this paper, an attempt is made to obtain the estimators for the parameters of one parameter Binomial type Rayleigh class Software Reliability Growth Model (SRGM) using Bayesian paradigm. The failure intensity of this model has been characterized by a mathematical function of total number of failures η and scale parameter η. The total numbers of failures present in the software initially and the rate at which the failures occurred before testing is unknown to the software tester due to this reason the non-informative priors are proposed for both the parameters. The estimators for the parameters η and η using Bayesian method have been obtained under squared error loss function. The performance of both the proposed Bayes estimators is compared with their related MLEs on the basis of risk efficiencies. The risk efficiencies are obtained by Monte Carlo simulation technique. It is observed that both the proposed Bayes estimators perform well for proper choices of execution time t.

Downloads

Download data is not yet available.