Christiane Lemieux

Monte Carlo and Quasi-Monte Carlo Sampling

The Monte Carlo method.- Sampling from known distributions.- Pseudorandom number generators.- Variance reduction techniques.- Quasi-Monte Carlo constructions.- Using quasi-Monte Carlo constructions.- Using quasi-Monte Carlo in practice.- Financial applications.- Beyond numerical integration.- Review of algebra.- Error and variance analysis for Halton sequences.- References.- Index.

Recent Advances in Randomized Quasi-Monte Carlo Methods

We survey some of the recent developments on quasi-Monte Carlo (QMC) methods, which, in their basic form, are a deterministic counterpart to the Monte Carlo (MC) method. Our main focus is the applicability of these methods to practical problems that involve the estimation of a high-dimensional integral. We review several QMC constructions and different randomizations that have been proposed to provide unbiased estimators and for error estimation. Randomizing QMC methods allows us to view them as variance reduction techniques. New and old results on this topic are used to explain how these methods can improve over the MC method in practice. We also discuss how this methodology can be coupled with clever transformations of the integrand in order to reduce the variance further. Additional topics included in this survey are the description of figures of merit used to measure the quality of the constructions underlying these methods, and other related techniques for multidimensional integration.

Extensible Lattice Sequences for Quasi-Monte Carlo Quadrature

Integration lattices are one of the main types of low discrepancy sets used in quasi-Monte Carlo methods. However, they have the disadvantage of being of fixed size. This article describes the construction of an infinite sequence of points, the first bm of which forms a lattice for any nonnegative integer m. Thus, if the quadrature error using an initial lattice is too large, the lattice can be extended without discarding the original points. Generating vectors for extensible lattices are found by minimizing a loss function based on some measure of discrepancy or nonuniformity of the lattice. The spectral test used for finding pseudorandom number generators is one important example of such a discrepancy. The performance of the extensible lattices proposed here is compared to that of other methods for some practical quadrature problems.

Quasi-Monte Carlo simulation of the light environment of plants

The distribution of light in the canopy is a major factor regulating the growth and development of a plant. The main variables of interest are the amount of photosynthetically active radiation (PAR) reaching different elements of the plant canopy, and the quality (spectral composition) of light reaching these elements. A light environment model based on Monte Carlo (MC) path tracing of photons, capable of computing both PAR and the spectral composition of light, was developed by Měch (1997), and can be conveniently interfaced with virtual plants expressed using the open L-system formalism. To improve the efficiency of the light distribution calculations provided by Měchs MonteCarlo program, we have implemented a similar program QuasiMC, which supports a more efficient randomised quasi-Monte Carlo sampling method (RQMC). We have validated QuasiMC by comparing it with MonteCarlo and with the radiosity-based CARIBU software (Chelle et al. 2004), and we show that these two programs produce consistent results. We also assessed the performance of the RQMC path tracing algorithm by comparing it with Monte Carlo path tracing and confirmed that RQMC offers a speed and/or accuracy improvement over MC.

Control Variates for Quasi-Monte Carlo

Quasi-Monte Carlo (QMC) methods have begun to displace ordinary Monte Carlo (MC) methods in many practical problems. It is natural and obvious to combine QMC methods with traditional variance reduction techniques used in MC sampling, such as control variates. There can, however, be some surprises. The optimal control variate coecient for QMC methods is not in general the same as for MC. Using the MC formula for the control variate coecient can worsen the performance of QMC methods. A good control variate in QMC is not necessarily one that correlates with the target integrand. Instead, certain high frequency parts or derivatives of the control variate should correlate with the corresponding quantities of the target. We present strategies for applying control variate coecients with QMC, and illustrate the method on a 16 dimensional integral from computational nance. We also include a survey of QMC aimed at a statistical readership.

Quasi-Regression and the Relative Importance of the ANOVA Components of a Function

In this paper we use quasi-regression to study high dimensional integrands. The variance of an integrand can be expressed as an infinite sum of squared components of an orthogonal basis. Various sums over subsets of these components have meaningful interpretations, and we develop numerical estimates for such sums. We pay particular attention to certain sums related to the ANOVA components of the integrand, because these are related to the effectiveness of quasi-Monte Carlo integration methods. We find that randomized quasi-Monte Carlo methods can estimate the unknown coefficients more accurately than ordinary Monte Carlo. We illustrate the method on two problems: valuing an Asian option, and finding the expected completion time in a stochastic activity network.

Randomized Polynomial Lattice Rules for Multivariate Integration and Simulation

Lattice rules are among the best methods to estimate integrals in a large number of dimensions. They are part of the quasi-Monte Carlo set of tools. A theoretical framework for a class of lattice rules defined in a space of polynomials with coefficients in a finite field is developed in this paper. A randomized version is studied, implementations and criteria for selecting the parameters are discussed, and examples of its use as a variance reduction tool in stochastic simulation are provided. Certain types of digital net constructions, as well as point sets constructed by taking all vectors of successive output values produced by a Tausworthe random number generator, are special cases of this method.

Acceleration of the Multiple-Try Metropolis algorithm using antithetic and stratified sampling

The Multiple-Try Metropolis is a recent extension of the Metropolis algorithm in which the next state of the chain is selected among a set of proposals. We propose a modification of the Multiple-Try Metropolis algorithm which allows for the use of correlated proposals, particularly antithetic and stratified proposals. The method is particularly useful for random walk Metropolis in high dimensional spaces and can be used easily when the proposal distribution is Gaussian. We explore the use of quasi Monte Carlo (QMC) methods to generate highly stratified samples. A series of examples is presented to evaluate the potential of the method.

Efficiency improvement by lattice rules for pricing Asian options

For the approximation of multidimensional integrals, two types of methods are widely used. Monte Carlo (MC) methods are the best known and require the use of a pseudorandom generator. Quasi-Monte Carlo (QMC) methods use low discrepancy point sets and are deterministic. The idea is to use points that are more regularly distributed over the integration space than random points. The best known methods to achieve this are the lattice rules and (t,s) sequences (or (t,m,s) nets) (A.B. Owen, 1998; H. Niederreiter; 1992; I.H. Sloan and S. Joe, 1994). The paper compares Monte Carlo methods, lattice rules, and other low discrepancy point sets on the problem of evaluating Asian options. The combination of these methods with variance reduction techniques is also explored.

On selection criteria for lattice rules and other quasi-Monte Carlo point sets

We define new selection criteria for lattice rules for quasi-Monte Carlo integration. The criteria examine the projections of the lattice over subspaces of small or successive dimensions. Their computation exploits the dimension-stationarity of certain lattice rules, and of other low-discrepancy point sets sharing this property. Numerical results illustrate the usefulness of these new figures of merit.