摘要翻译:
本文给出了形式为$q(x)=f^{-1}(G(x))$的函数的微分方程和求解方法,其中$f$和$G$是累积分布函数。这种功能允许直接回收蒙特卡罗样本从一个分布到另一个样本。对于某些特殊情况,该方法可以进行分析,并说明它是传统康沃什-费雪展开式的一种更精确的形式。以这种方式,可以在没有与重采样相关联的蒙特卡罗噪声的情况下评估分布风险的模型风险。给出了将正态样本转换为学生t,将指数样本转换为双曲型、方差γ和正态样本的方程。在正态分布的情况下,所采用的变量的变化允许在很宽的样本空间范围内基于单一有理近似以很好的精度进行抽样。避免任何分支语句在最佳GPU计算中是有用的,因为它避免了{it warp发散}的影响,我们给出了在GPU环境中提供性能改进的无分支正规分位数的例子,同时保留了众所周知的方法的最佳精度特征。我们还提供了基于低概率偏差的模型。在Nvidia Quadro 4000、GTX285和480以及Tesla C2050 GPU上进行了新旧形式的比较。我们认为,在单精度模式下,变量变化方法提供了与现有最快方案竞争的性能,同时大幅提高了精度;在双精度模式下,该方法提供了迄今为止GPU最优的高斯分位数,并且不影响蒙特卡罗应用的精度,工作速度是CUDA 4库函数的两倍,提高了精度。
---
英文标题:
《Quantile Mechanics II: Changes of Variables in Monte Carlo methods and
GPU-Optimized Normal Quantiles》
---
作者:
William T. Shaw, Thomas Luu, Nick Brickman
---
最新提交年份:
2011
---
分类信息:
一级分类:Quantitative Finance 数量金融学
二级分类:Computational Finance 计算金融学
分类描述:Computational methods, including Monte Carlo, PDE, lattice and other numerical methods with applications to financial modeling
计算方法,包括蒙特卡罗,偏微分方程,格子和其他数值方法,并应用于金融建模
--
一级分类:Quantitative Finance 数量金融学
二级分类:Risk Management 风险管理
分类描述:Measurement and management of financial risks in trading, banking, insurance, corporate and other applications
衡量和管理贸易、银行、保险、企业和其他应用中的金融风险
--
一级分类:Quantitative Finance 数量金融学
二级分类:Statistical Finance 统计金融
分类描述:Statistical, econometric and econophysics analyses with applications to financial markets and economic data
统计、计量经济学和经济物理学分析及其在金融市场和经济数据中的应用
--
一级分类:Statistics 统计学
二级分类:Applications 应用程序
分类描述:Biology, Education, Epidemiology, Engineering, Environmental Sciences, Medical, Physical Sciences, Quality Control, Social Sciences
生物学,教育学,流行病学,工程学,环境科学,医学,物理科学,质量控制,社会科学
--
一级分类:Statistics 统计学
二级分类:Computation 计算
分类描述:Algorithms, Simulation, Visualization
算法、模拟、可视化
--
---
英文摘要:
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.
---
PDF链接:
https://arxiv.org/pdf/0901.0638