ABSTRACT
This chapter describes the construction of statistical tolerance limits for a continuous random variable. Statistical tolerance limits estimate the range of values that contain a specified percentage of a population with a given level of confidence. Both two-sided tolerance intervals and one-sided tolerance bounds are included. Several types of limits are considered: limits for data from a normal distribution, limits for data that is normally distributed after a transformation, limits for data from nonnormal distributions such as the largest extreme value distribution, and nonparametric tolerance limits that make no assumptions about distributional form.
