Hands-On Machine Learning for Algorithmic Trading

Markov chain Monte Carlo sampling

A Markov chain is a dynamic stochastic model that describes a random walk over a set of states, connected by transition probabilities. The Markov property stipulates that the process has no memory, and the next step only depends on the current state. In other words, it's conditional on the present, past, and future being independent, that is, information about past states does not help to predict the future beyond what we know from the present.

Monte Carlo methods rely on repeated random sampling to approximate results that may be deterministic, but that does not permit an analytic, exact solution. It was developed during the Manhattan Project to estimate energy at the atomic level and received its enduring ...