Bootstrap methods are procedures for the empirical estimation or approximation of sampling distributions and their characteristics. Their primary use lies in the estimation of accuracy measures, such as bias and variance, for parameter estimators, and in construction of confidence sets or hypothesis tests for population parameters. They are applied in circumstances in which the form of the population from which the observed data have been drawn is unknown. They prove particularly useful where very limited sample data are available and traditional parametric modeling and analysis are difficult or unreliable.