版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领
文档简介
1、Monte Carlo Simulations using MATLABVincent Leclercq, Application engineerEmail : vincent.leclercqmathworks.frAgendaPrinciples and uses cases for Monte Carlo methodsUsing MATLAB toolbox for Monte Carlo simulationsDevelop you own Monte Carlo engineA quick overview of Variance reduction technicsAn sim
2、ple exampleCompute the area of a lakeWe shot N cannon ballsn balls out of the “lake” N- n balls in the lakeOne getsTake outsResults depends on :random number generation (Mersenne Twister)Number of simulationsTypical uses cases of Monte carlo in financeDerivatives pricingRiskStructurerStochastic Asse
3、t Liability ManagementGeneral PrinciplesEstimation technique based on the simulation of a great number of random variablesLets consider This can be seen as the expectation , with U being a uniform random variable on (0,1), ie U(0,1)We can estimate tis expectation using an empiric mean from random dr
4、awsGeneral Principles, continuedWe need to generate a sequence Ui of random samples, which are independent (iid)Then we can compute the empiric mean:From the law of Great Numbers, one gets :Variance :Confidence interval : using a Gaussian approximation Good points and drawbacksGood pointsVarious app
5、lication areasFew hypothesisEasy to developDrawbacksDependency to random number generatorBig variability (accuracy)Computation timeWhy MATLAB ?Efficiency :1 000 000 paths in less than 1 s25 times fastest thanExcelState of the Art AlgorithmsMersenne TwisterLinear algebraLots of statistical distributi
6、ons supported (+ than 20)Easy deploymentAgendaPrinciples and uses cases for Monte Carlo methodsUsing MATLAB toolbox for Monte Carlo simulationsDevelop you own Monte Carlo engineA quick overview of Variance reduction technicsWhich tools for Monte Carlo simulations ?MATLAB : Core linear algebra engine
7、, matrix factorisation, Statistics toolbox : Random numbers, copulas, Financial toolbox :Portsim :” Monte Carlo simulation of correlated asset returns”GARCH ToolboxgarchsimFinancial toolbox : portsimOn a time interval, performances are driven by the following equation :Time basis must be consistent
8、for input parameters (drift and volatility)Annual time basis - dt in yearsDaily time basis- dt in daysDemo 1:Geometric brownian motion Lognormality of equity pricesHistorical data inputDrift and volatilityAnnually or Daily Simulate 10 000 paths on one year.Compare resultsDemo 2: Use the previous pat
9、hs to price aVanilla optionApply the option payoffVanilla - No path dependancyCompute the call price for different strikesCompute the confidence intervalsGARCH Toolbox : garchsimStochastic VolatilitySimulations of Auto Regressive models / GARCH“Perform Monte Carlo simulation of univariate returns, i
10、nnovations,and conditional volatilities”Fitting (Adjust the model, garchfit function) and SimulationSimulation , several possibilities:Use of historic data (bootstrapping)See Market Risk Using Bootstrapping and Filtered Historical SimulationUse of random variablesAgendaPrinciples and uses cases for
11、Monte Carlo methodsUsing MATLAB toolbox for Monte Carlo simulationsDevelop you own Monte Carlo engineA quick overview of Variance reduction technicsWhat do I need for Monte Carlo ?A good random number generatorRand, randn - several chocie possible for random number generationRandom (more than 20 dis
12、tributions), copularnd - Statistics toolboxLinear algebra functions: Cholesky factorizationcumsumProcessGenerate Random numbersDirectly from the statistical distributionThrough a uniform law- Allow the use of quasi random number generationApply the model (volatility, )Computation of the empiric mean
13、Confidence interval estimationDemoCorrelated Equities SimulationInput :Time : NDaysNumber of different paths : NSimulation Number of Assets : 2, NAssets with correlationWe know :VolatilityCorrelationsOutput :Matrice de NDays* NSimulation*NAssetsPreserved CorrelationsAgendaPrinciples and uses cases f
14、or Monte Carlo methodsUsing MATLAB toolbox for Monte Carlo simulationsDevelop you own Monte Carlo engineA quick overview of Variance reduction technicsVariance ReductionWhy ?Slow Convergence of Monte Carlo pricingNeed a great number of pathsSolution : Use if various variance reduction methodsSeveral
15、 possible methodsVariance Reduction : OverviewAntithetic VariablesEfficient, easy to implementEfficiency depends of the option (ex : Butterfly)Control VariablesUse of a variable correlated to the one we want to estimateEx : Vanilla option PricingWe canuse the close formula (Hulll) in order to comput
16、e the variance and the expected return of the underlying at maturityWe need to estimate the covariance between our control variable (the underlying) and the variable we want to estimate (option price)Variance Reduction Overview (2/3)Quasi Monte CarloUse of low discrepancy sequences“quasi random” seq
17、uencesHalton sequences, Sobol sequences, Better AccuracyVariance Reduction Overview (3/3)Variance reduction using conditionning Principle:Var(EX) Reduced VarianceOther techniques :Importance samplingStratified samplingDemonstrationVanilla option pricing using Variance ReductionSeveral methodology us
18、edAntithetic VariablesQuasi Monte Carlo (Halton / Sobol)Control VariableResults comparisonVariance Reduction, Key takeoutsEfficient, Generic methodConfidence intervalsVariance Reduction technics should be used wisely, depending on the product to priceExample : Antithetic for options Butterfly lead to an increase of the varianceLots of research papersGeneral ConclusionMATLAB allow users to quickly develop and test advanced Monte Carlo simulation Very generic solutionNew : a complete framework Monte Carlo
温馨提示
- 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
- 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
- 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
- 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
- 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
- 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
- 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。
最新文档
- 2026中国新能源汽车电机驱动系统行业市场深度调研及发展前景与投资分析
- 老化房设备操作和维护保养规程
- 2026冷链物流运输箱体铰链衬套耐候性改良与成本优化方案
- 2026中国智能交通系统行业市场供需关系与投资可行性
- 2026中国音响制造行业市场现状供需分析及投资评估规划分析研究报告
- 2026运动护腰产品临床效果验证与营销策略
- 2026中国无人零售终端市场消费行为与商业价值研究报告
- 2026中国通信设备行业市场需求技术标准竞争分析投资评估规划研究报告
- 2026中国智能服装行业市场现状供需分析投资评估发展规划报告
- 2026轻工业品行业现状供需分析及投资评估规划分析研究报告
- GA/T 1043-2025智能交通管理系统前端设备运行维护规范
- (高清版)DB31∕T 1083-2025 公共停车信息联网技术要求
- 石油化工设备维护检修规程设备完好标准SHS010012023年
- 供应商来料包装规范-
- 南充市医疗保险特殊门诊申请表
- 装饰公司绩效考核岗位职责
- 七年级开学第一课课件
- GB∕T 2567-2021 树脂浇铸体性能试验方法
- Q∕SY 04797-2020 燃油加油机应用规范
- 2019译林版高中英语必修三单词默写表
- 恢复驾驶资格科目一考试题库(450题)
评论
0/150
提交评论