Simulation-based interval estimation approach in software reliability assessment
Shinji Inoue, Shigeru Yamada · 2016
It is known that it is almost impossible to derive a probability distribution for parameters in nonhomogeneous Poisson process software reliability growth models and the conventional point estimation method leads mismatched estimation when the number of sampling data is not sufficiently large. For overcoming this problems, we discuss a nonparametric bootstrap method for interval estimations of software reliability and cost-optimal software release time by using a discretized software reliability growth model. And we discuss three types of bootstrap confidence interval estimation methods, such as basic, standard normal, and percentile bootstrap confidence interval estimation methods, for interval estimations of model parameters, several software reliability assessment measures, and cost-optimal software release time.