Lectures on Monte Carlo methods (Q5952635): Difference between revisions

This book is based on the notes for a course the author gave at the Fields Institute for the Research in Mathematical Sciences in Fall 1998. The aim of the course, and of the book as well, was to introduce students (readers) to the theory and practice of Monte Carlo methods. The course (and the book) was prepared not only for pure mathematicians, but also for people whose research in statistics, physics, chemistry, finance, computer science and biology. Therefore, the only formal prerequisite for the course is an undergraduate course in probability, although it is also helpful to know some basic statistics and a knowledge of elementary Markov chains theory is assumed for Chapter~4. In any case, the level of the book is suitable for a beginning graduate student or an advanced undergraduate. The first chapter contains an overview of what Monte Carlo is all about, as well as a brief summary of basic concepts from probability and statistics that the reader is assumed to have seen elsewhere. Chapter~2 describes various algorithms that produce ``random numbers'' having desired distribution. Computer implementation of these algorithms are the fundamental building blocks of any Monte Carlo study. Chapter~3 describes techniques for improving the efficiency and accuracy of Monte Carlo studies, and illustrate the concepts with running examples from numerical integration, network reliability and queuing theory. Chapter~4 is devoted to the MCMC, a powerful method for simulating complex random systems arising in statistics, physics, computer science, and other fields. Chapter~5 describes statistical methods for analysing the output of a Monte Carlo study. Chapter~6 concludes the book with a discussion of the Ising model from statistical physics and MCMC methods that have been developed to study it.