Abstract:
We provide a pedagogical introduction to the two main variants of real-space quantum Monte Carlo methods for electronic-structure calculations: variational Monte Carlo (VMC) and diffusion Monte Carlo (DMC). Assuming no prior knowledge on the subject, we review in depth the Metropolis-Hastings algorithm used in VMC for sampling the square of an approximate wave function, discussing details important for applications to electronic systems. We also review in detail the more sophisticated DMC algorithm within the fixed-node approximation, introduced to avoid the infamous Fermionic sign problem, which allows one to sample a more accurate approximation to the ground-state wave function. Throughout this review, we discuss the statistical methods used for evaluating expectation values and statistical uncertainties. In particular, we show how to estimate nonlinear functions of expectation values and their statistical uncertainties.

Abstract:
In these lectures, given in '96 summer schools in Beg-Rohu (France) and Budapest, I discuss the fundamental principles of thermodynamic and dynamic Monte Carlo methods in a simple light-weight fashion. The keywords are MARKOV CHAINS, SAMPLING, DETAILED BALANCE, A PRIORI PROBABILITIES, REJECTIONS, ERGODICITY, "FASTER THAN THE CLOCK ALGORITHMS". The emphasis is on ORIENTATION, which is difficult to obtain (all the mathematics being simple). A firm sense of orientation helps to avoid getting lost, especially if you want to leave safe trodden-out paths established by common usage. Even though I remain quite basic (and, I hope, readable), I make every effort to drive home the essential messages, which are easily explained: the crystal-clearness of detail balance, the main problem with Markov chains, the great algorithmic freedom, both in thermodynamic and dynamic Monte Carlo, and the fundamental differences between the two problems.

Abstract:
The author's presentation of multilevel Monte Carlo path simulation at the MCQMC 2006 conference stimulated a lot of research into multilevel Monte Carlo methods. This paper reviews the progress since then, emphasising the simplicity, flexibility and generality of the multilevel Monte Carlo approach. It also offers a few original ideas and suggests areas for future research.

Abstract:
Monte Carlo methods are now an essential part of the statistician's toolbox, to the point of being more familiar to graduate students than the measure theoretic notions upon which they are based! We recall in this note some of the advances made in the design of Monte Carlo techniques towards their use in Statistics, referring to Robert and Casella (2004,2010) for an in-depth coverage.

Abstract:
This article reviews the basic computational techniques for carrying out multi-scale simulations using statistical methods, with the focus on simulations of epitaxial growth. First, the statistical-physics background behind Monte Carlo simulations is briefly described. The kinetic Monte Carlo (kMC) method is introduced as an extension of the more wide-spread thermodynamic Monte Carlo methods, and algorithms for kMC simulations, including parallel ones, are discussed in some detail. The step from the atomistic picture to the more coarse-grained description of Monte Carlo simulations is exemplified for the case of surface diffusion. Here, the aim is the derivation of rate constants from knowledge about the underlying atomic processes. Both the simple approach of Transition State Theory, as well as more recent approaches using accelerated molecular dynamics are reviewed. Finally, I address the point that simplifications often need to be introduced in practical Monte Carlo simulations in order to reduce the complexity of 'real' atomic processes. Different 'flavors' of kMC simulations and the potential pitfalls related to the reduction of complexity are presented in the context of simulations of epitaxial growth.

Abstract:
In low-temperature high-density plasmas quantum effects of the electrons are becoming increasingly important. This requires the development of new theoretical and computational tools. Quantum Monte Carlo methods are among the most successful approaches to first-principle simulations of many-body quantum systems. In this chapter we present a recently developed method---the configuration path integral Monte Carlo (CPIMC) method for moderately coupled, highly degenerate fermions at finite temperatures. It is based on the second quantization representation of the $N$-particle density operator in a basis of (anti-)symmetrized $N$-particle states (configurations of occupation numbers) and allows to tread arbitrary pair interactions in a continuous space. We give a detailed description of the method and discuss the application to electrons or, more generally, Coulomb-interacting fermions. As a test case we consider a few quantum particles in a one-dimensional harmonic trap. Depending on the coupling parameter (ratio of the interaction energy to kinetic energy), the method strongly reduces the sign problem as compared to direct path integral Monte Carlo (DPIMC) simulations in the regime of strong degeneracy which is of particular importance for dense matter in laser plasmas or compact stars. In order to provide a self-contained introduction, the chapter includes a short introduction to Metropolis Monte Carlo methods and the second quantization of quantum mechanics.

Abstract:
Monte Carlo simulation with {\it a-priori} unknown weights have attracted recent attention and progress has been made in understanding (i) the technical feasibility of such simulations and (ii) classes of systems for which such simulations lead to major improvements over conventional Monte Carlo simulations. After briefly sketching the history of multicanonical calculations and their range of application, a general introduction in the context of the statistical physics of the d-dimensional generalized Potts models is given. Multicanonical simulations yield canonical expectation values for a range of temperatures or any other parameter(s) for which appropriate weights can be constructed. We shall address in some details the question how the multicanonical weights are actually obtained. Subsequently miscellaneous topics related to the considered algorithms are reviewed. Then multicanonical studies of first order phase transitions are discussed and finally applications to complex systems such as spin glasses and proteins.

Abstract:
I discuss optimized data analysis and Monte Carlo methods. Reweighting methods are discussed through examples, like Lee-Yang zeroes in the Ising model and the absence of deconfinement in QCD. I discuss reweighted data analysis and multi-hystogramming. I introduce Simulated Tempering, and as an example its application to the Random Field Ising Model. I illustrate Parallel Tempering, and discuss some technical crucial details like thermalization and volume scaling. I give a general perspective by discussing Umbrella Methods and the Multicanonical approach.

Abstract:
A self-contained and tutorial presentation of the diffusion Monte Carlo method for determining the ground state energy and wave function of quantum systems is provided. First, the theoretical basis of the method is derived and then a numerical algorithm is formulated. The algorithm is applied to determine the ground state of the harmonic oscillator, the Morse oscillator, the hydrogen atom, and the electronic ground state of the H2+ ion and of the H2 molecule. A computer program on which the sample calculations are based is available upon request.

Abstract:
Monte Carlo is a simple and flexible tool that is widely used in computational finance. In this context, it is common for the quantity of interest to be the expected value of a random variable defined via a stochastic differential equation. In 2008, Giles proposed a remarkable improvement to the approach of discretizing with a numerical method and applying standard Monte Carlo. His multilevel Monte Carlo method offers an order of speed up given by the inverse of epsilon, where epsilon is the required accuracy. So computations can run 100 times more quickly when two digits of accuracy are required. The multilevel philosophy has since been adopted by a range of researchers and a wealth of practically significant results has arisen, most of which have yet to make their way into the expository literature. In this work, we give a brief, accessible, introduction to multilevel Monte Carlo and summarize recent results applicable to the task of option evaluation.