Most standard distributions involve constants whose values are unknown — unknown parameters. This chapter describes some methods for estimating the values of such parameters using a random sample from the distribution.
Two general estimation methods are called the method of moments and maximum likelihood. Maximum likelihood can be used for a wider range of models and parameters and some large-sample results make it particularly useful.
The maximum likelihood estimate from a single sample is usually not exactly equal to the unknown parameter value. Confidence intervals describe uncertainty in the estimate and a general method to find confidence intervals for maximum likelihood estimates is described.