互動式金融與風險量化計算器 · E-E-A-T 公式審計
ZH-HANT · 繁體中文風險值(VaR)
Value at Risk estimates the loss a position should not exceed over a given horizon at a chosen confidence level. Expected Shortfall is the average loss in the cases where that threshold is breached.
Value at Risk
83,415.53
8.34% of the position
- Expected Shortfall
- 95,566.21
- Volatility over horizon
- 3.59%
1. 核心數學公式與定價恆等式
σ_horizon = σ_annual × √(days ÷ 252)
VaR = Position × z_α × σ_horizon
Expected Shortfall = Position × σ_horizon × φ(z_α) ÷ (1 − α)
2. 參數輸入指南與使用方法
Enter the position size, its annualised volatility, the confidence level and the holding horizon in trading days.
Volatility is scaled to the horizon with the square-root-of-time rule, then multiplied by the normal quantile for the confidence level.
3. 模型審慎假設與適用邊界限制
- Assumes normally distributed returns with zero mean. Real returns have fatter tails, so parametric VaR tends to understate extreme losses.
- Single position only; for a portfolio you would need the full covariance matrix.
4. 常見問題與量化風控解答(FAQ)
Why show Expected Shortfall as well as VaR?
VaR says nothing about how large losses are once the threshold is crossed. Expected Shortfall does, which is why Basel market-risk rules moved to a 97.5% Expected Shortfall measure.
Why 252 days?
252 is the usual number of trading days in a year. Using calendar days would understate daily volatility.
如何在機構風控工作流中驗證「風險值(VaR)」的計算結果?
QuantCalc Tools 在本頁完整列出了該工具的閉式數學公式與計息基準假設。風控與合規團隊可將相同參數代入 Python/Pandas 或終端風控閘道進行獨立複核,並透過 QuantCalc Tools 牌照核驗中心確認交易對手方的監管授權狀態。
免責聲明:本工具僅供金融工程研究與教育用途,不構成任何投資或交易建議。計算結果完全取決於您輸入的參數及上述模型假設,重要財務決策前請獨立核實。
