In statistics, boostrap is a Monte Carlo technique to estimate the Sampling distribution of an Estimator

Parametric bootstrap

Uses estimated parameter to generate samples

Nonparametric bootstrap

can refer to any test or metric that relies on random sampling with replacement. This technique allows estimation of the sampling distribution of almost any statistic using random sampling methods.

https://www.wikiwand.com/en/Bootstrapping_(statistics)

• Bootstrap aggregation is an example.

Explanation: http://stats.stackexchange.com/questions/26088/explaining-to-laypeople-why-bootstrapping-works

An Introduction to the Bootstrap