ABSTRACT
The gamma distribution is a classical distribution which has appeared in the literature since the early 1800s. The gamma distribution is studied as a population model or "parent" distribution. The Weibull distribution was a generalization of the exponential distribution, and the gamma distribution provides an alternate generalization. Certainly the gamma distribution may be a useful model for many other types of characteristics and situations, and statistical techniques would then be applicable. If failed parts are replaced with new parts when they fail, then the time between failures of a system may reasonably follow the gamma distribution. The best approach to maximum likelihood estimation for the gamma distribution is the use of the rational approximation. The chapter discusses the method of maximum likelihood for point estimation of the parameters. Estimation of the parameters becomes considerably more difficult under censored sampling.
