VaR模型 Value at Risk (VaR) Model - 用户指南

Value-at-Risk (VaR) is a common measure of the degree of risk a portfolio is exposed to, by calculating how much an investor might lose with his/her portfolio, given a certain probability, within a certain time period. For example, a 1-day VaR of $1000 with 95% confidence interval implies that the portfolio might lose $1000 or more with 5% probability in 1 day.

风险价值 (VaR) 是衡量交易组合所面临风险程度的常用指标，通过计算投资者在给定一定概率下在特定时间段内的交易组合可能损失多少。例如，1 天的 VaR 为 1000 美元，置信区间为 95%，这意味着投资组合可能会在 1 天内以 5% 的概率损失 1000 美元或更多。

The Value-at-Risk model simulates a large number of outcomes of the portfolio's performance, based on the assets' past returns and volatility data, and displays the portfolio's projected profit or loss (P/L) as a probability distribution. The amount of P/L is shown on the horizontal axis (X-axis), while the likelihood measures are shown on the vertical axis (Y-axis). 2 plots are created: the probability density bar chart shows the frequencies of the various P/L outcomes, while the cumulative distribution graph shows the probability of losing that amount or more.

风险价值模型根据资产过去的收益和波动率数据模拟投资组合业绩的大量结果，并将投资组合的预计损益 (P/L) 显示为概率分布。 P/L 的数量显示在水平轴（X 轴）上，而可能性度量显示在垂直轴（Y 轴）上。创建了 2 个图：概率密度条形图显示各种 P/L 结果的频率，而累积分布图显示损失该金额或更多的概率。

Losses beyond a certain threshold (ie. probability of loss < 1 - confidence interval) are highlighted on the left tail of the graph. Furthermore, 2 values are calculated and displayed: VaR is the threshold value, while shortfall value is the probability-weighted average of the tail losses.

超过某个阈值的损失（即损失概率 < 1 - 置信区间）在图表的左尾突出显示。此外，计算并显示了 2 个值：VaR 是阈值，而 shortfall 值是尾部损失的概率加权平均值。

Below Shortfall, Add Sample to show Size 样本数