ABSTRACT
This chapter is concerned with truncation and censoring in the exponential distribution. This distribution is an important model in life time and reaction time studies. Truncation on the left is of little importance in life time studies, although it might be of interest in other applications. Censoring on the right occurs in life testing and related studies when sample items are removed from a test before all have reacted. Censoring may occur in a single stage or it may be progressive and occur in multiple stages. Hybrid censoring is a mixture of Type I and Type II censoring. Samples from truncated distributions are referred to as truncated samples. They differ from censored samples in that the number of missing observations is an unknown quantity. B. Epstein and M. Sobel noted relationships that exist between estimates of exponential parameters from censored samples and the chi-square distribution.
