On Discrete Failure Models.
Abstract
In some situations, discrete failure time distributions are more appropriate to model lifetimes than continuous distributions. Very few results for the discrete case have been given in the literature. This paper provides three families of discrete parametric lifetime distribution which are quite versatile in fitting Increasing Failure Rate (IFR) and Decreasing Failure Rate (DFR), and constant failure rate models to either uncensored or right-censored life-test data. Teh maximum likelihood estimation (MLE) of parameters, survival probabilities, and mean lifetimes is investigated, and the MLEs are shown to be easily computed by simple numerical methods. An example is given for each of the models, allowing the comparison of the proposed models. The example illustrates that the discrete models presented can provide a better fit to discrete data than previously proposed discrete distributions.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 17, 1984
- Accession Number
- ADA145579
Entities
People
- J. D. Spurrier
- William J. Padgett
Organizations
- University of South Carolina