Most introductory statistics courses deal with data that are assumed to be a random sample from some numerical or categorical distribution. For models of this type, statistical theory gives the properties of sample means, proportions, etc, and allows us to perform inference.
Many real-life situations may be approximated by models whose descriptions are equally simply but whose consequences cannot be easily determined from statistical theory. This module describes how simulations can be used to help analyse such models and to assess whether they are adequate descriptions of reality.
The main pages in this module can be understood with minimal statistical background — an understanding of the basic concepts of probability and normal distributions suffice. However there are several optional linked pages that either provide further detail on topics or give an extended analysis of the simulation results. Some familiarity with statistical inference and rectangular and binomial distributions is needed to fully understand this additional material. The optional topics include:
This module is part of the CAST collection of e-books. Readers who have never met probability or normal distributions are advised to study the pages about sampling phenomena in a CAST introductory e-book before attempting this module.