Quantile mechanics II: Changes of variables in Monte Carlo methods and GPU-optimised normal quantiles

DOI10.1017/S0956792513000417MaRDI QIDQ2878024zbMATH OpenOpenAlexFDO

Authors William T. Shaw, Thomas Luu, Nick Brickman

Publication date 28 August 2014

Published in European Journal of Applied Mathematics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0901.0638

zbMATH Keywords

probitCUDAcomputational financeGaussianinverse CDF

Mathematics Subject Classification ID

Exact distribution theory in statistics (62E15) Asymptotic distribution theory in statistics (62E20) Applications of statistics to actuarial sciences and financial mathematics (62P05) Numerical methods (including Monte Carlo methods) (91G60) Statistical methods; risk measures (91G70) General considerations in statistical decision theory (62C05)

Abstract: This article presents differential equations and solution methods for the functions of the form $Q(x)=F^{-1}(G(x))$, where $F$ and $G$ are cumulative distribution functions. Such functions allow the direct recycling of Monte Carlo samples from one distribution into samples from another. The method may be developed analytically for certain special cases, and illuminate the idea that it is a more precise form of the traditional Cornish-Fisher expansion. In this manner the model risk of distributional risk may be assessed free of the Monte Carlo noise associated with resampling. Examples are given of equations for converting normal samples to Student t, and converting exponential to hyperbolic, variance gamma and normal. In the case of the normal distribution, the change of variables employed allows the sampling to take place to good accuracy based on a single rational approximation over a very wide range of the sample space. The avoidance of any branching statement is of use in optimal GPU computations as it avoids the effect of {it warp divergence}, and we give examples of branch-free normal quantiles that offer performance improvements in a GPU environment, while retaining the best precision characteristics of well-known methods. We also offer models based on a low-probability of warp divergence. Comparisons of new and old forms are made on the Nvidia Quadro 4000, GTX 285 and 480, and Tesla C2050 GPUs. We argue that in single-precision mode, the change-of-variables approach offers performance competitive with the fastest existing scheme while substantially improving precision, and that in double-precision mode, this approach offers the most GPU-optimal Gaussian quantile yet, and without compromise on precision for Monte Carlo applications, working twice as fast as the CUDA 4 library function with increased precision.

Recommendations

• Quantile mechanics
• Efficient and accurate parallel inversion of the gamma distribution
• Efficient numerical inversion for financial simulations
• Rapid evaluation of the inverse of the normal distribution function
• High-performance financial simulation using randomized quasi-Monte Carlo methods

Cites work

• scientific article; zbMATH DE number 1999206 (Why is no real title available?)
• Copulas: Tales and facts (with discussion)
• Generalized Asymptotic Expansions of Cornish-Fisher Type
• On the Distributional Characterization of Daily Log‐Returns of a World Stock Index

Cited in

(2)

• Efficient and accurate parallel inversion of the gamma distribution
• Efficient estimation of financial risk by regressing the quantiles of parametric distributions: an application to CARR models

Describes a project that uses

Uses Software

• CUDA

This page was built for publication: Quantile mechanics II: Changes of variables in Monte Carlo methods and GPU-optimised normal quantiles

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2878024)