第14章-风险价值VaR计算.pptx

第14章 风险价值VaR计算;14.1　VaR模型;14.1.1 　VaR模型的含义;14.2　VaR计算方法;14.3.1 　数据提取;14.3.2 　数据可视化与标准化;%选定股票价格序列 mypickStockPrices = CSI300HistPrices(:,mypick); %选定股票的标准价格 mypickNormPrices = normPrices(:,mypick); %选定股票的名称 mypickCSI300Tickers = CSI300Tickers(mypick); %绘制图形 plot(mypickNormPrices,DisplayName,mypickNormPrices,YDataSource,mypickNormPrices);figure(gcf) %添加图示 legend(mypickCSI300Tickers) %指数标准价格 normIndexPrice = ret2tick(tick2ret(PortfoliopricesIndex)); %在上图中添加指数曲线 hold all plot(normIndexPrice,DisplayName,Index,YDataSource,normIndexPrice);figure(gcf);14.3.2 　数据可视化与标准化;14.3.3 　数据简单处理与分析;14.3.3 　数据简单处理与分析;14.4　数据处理;投资组合净值与收益率分布;14.5　历史模拟法程序;;14.6　参数模型法程序;代码如下：;;%% Monte Carlo using portsim %蒙特卡罗方法 % 从文件CSI300.xlsx的CSI300中读取数据 [num,txt]=xlsread(CSI300.xlsx,CSI300); CSI300HistPrices=num;%成分股历史价格 [num,txt]=xlsread(CSI300.xlsx,2); positionsPortfolio=num;%positionsPortfolio 股票数量 %根据价格序列计算收益率 returnsSecurity = tick2ret(CSI300HistPrices,[],continuous); %根据组合中股票价格与股票数量,计算组合资产价值与权重 [marketValuePortfolio,weightsPortfolio]= getPortfolioWeights(... CSI300HistPrices,positionsPortfolio); %具体参见getPortfolioWeights程序;;;;%% 使用GBM 对象进行蒙特卡罗模拟 expReturn = mean(returnsSecurity); sigma = std(returnsSecurity); correlation = corrcoef(returnsSecurity); X = CSI300HistPrices(end,:); dt = 1; numObs = 1;% Numberofobservation numSim = 10000;% Numberofsimulation rng(12345) GBM = gbm(diag(expReturn),diag(sigma),Correlation,correlation,StartState,X);;% SimulatefornumSimtrials simulatedAssetPrices = GBM.simulate(numObs,DeltaTime,dt,ntrials,numSim); simulatedAssetReturns = tick2ret(simulatedAssetPrices,[],continuous); % simulatedAssetReturns = squeeze(simulatedAssetReturns); simulatedAssetReturns = exp(squeeze(simulatedAssetReturns))-1; gbmVals = weightsPortfolio*simulatedAssetReturns; gbmVar = -prctile(gbmVals*marketValuePortfolio,[1 5]); % Visualizethesimulatedportfolios plotMonteCarlo(gbmVals) % ValueatRisk displayVar(gbmVar(1),gbmVar(2),mcg);蒙特卡罗刹模拟方法(GBM)结果;