Abstract by Steven Orgill
Parameter Estimation of the Weibull Distribution
The Weibull distribution is widely used in a variety of settings. The purpose of this project is to compare different parameter estimation techniques for the Weibull distribution. Maximum likelihood, method of moments, and Bayesian estimate under squared error loss are used to estimate the shape and scale parameters from a known Weibull distribution. Understanding how effective these estimation techniques are can then inform decisions on trying to fit a Weibull distribution to data with an unknown parameterization. As an example these methods are applied to estmate parameters for a Weibull distribution to model MLB free agent salary data.