Study guide
11.2 Measuring and Managing Market Risk: VaR, CVaR & Stress Testing
CFA Level 2 Chartered Financial Analyst
Track your progress as you read
Finished sections get a check mark.
1 Introduction & CFA Level II Exam Strategy
1.1 CFA Level II Exam Overview & Item-Set Vignette Strategy 1.2 2026 Curriculum Structure & 10 Topic Weight Bands 1.3 300-Hour Study Plan, Calculator Mastery & Practical Skills Modules (PSM)
2 Ethical and Professional Standards
2.1 Code of Ethics & The Seven Standards of Professional Conduct 2.2 Vignette Analysis: Conflicts of Interest & Disclosure (Standard VI) 2.3 Duties to Clients (Standard III) & Duties to Employers (Standard IV) 2.4 Standard V: Investment Analysis, Recommendations & Actions and Standard VII: Responsibilities as a CFA Member or Candidate 2.5 Application of the Code and Standards: Level II Case Method 2.6 Recommended Procedures: Compliance Programs, Supervisory Systems & Firm Policies
3 Quantitative Methods
3.1 Multiple Linear Regression, ANOVA & Hypothesis Testing 3.2 Regression Diagnostics: Multicollinearity, Heteroskedasticity & Serial Correlation 3.3 Time-Series Analysis: AR Models, Nonstationarity & Cointegration 3.4 Machine Learning, Deep Learning & Big Data Projects 3.5 Extensions of Multiple Regression: Influence Analysis, Qualitative Variables & Logistic Regression 3.6 Big Data Projects: Wrangling, Exploration, Feature Engineering & Model Training
4 Economics for Valuation
4.1 Foreign Exchange Parity Relations & Equilibrium Exchange Rates 4.2 Economic Growth Theories & Production Function Analysis 4.3 Forward Points, the Carry Trade & Policy Effects on Exchange Rates 4.4 Growth Determinants: Natural Resources, Demographics, Convergence & Trade Openness
5 Financial Statement Analysis
5.1 Intercorporate Investments: Equity Method & Business Combinations 5.2 Employee Compensation: Defined Benefit Pension Plans & Share-Based Payments 5.3 Multinational Operations: Current Rate vs Temporal Translation 5.4 Analysis of Financial Institutions & The CAMELS Framework 5.5 Evaluating the Quality of Financial Reports & Accounting Adjustments 5.6 Integration of Financial Statement Analysis Techniques
6 Corporate Issuers
6.1 Analysis of Dividends: Policy Theories, Signaling & Taxation Systems 6.2 Share Repurchases: Methods, EPS and Book Value Effects & Payout Sustainability 6.3 Cost of Capital: Advanced Topics 6.4 Corporate Governance, Board Oversight & ESG Integration 6.5 Corporate Restructuring: Types, Motivations, Evaluation & Valuation 6.6 Restructuring Financial Effects: EPS, Net Debt to EBITDA, WACC & Divestitures
7 Equity Valuation
7.1 Equity Valuation Framework & Required Rate of Return Estimation 7.2 Discounted Dividend Valuation: Gordon Growth, Multistage & The H-Model 7.3 Free Cash Flow Valuation: FCFF and FCFE Modeling 7.4 Market-Based Valuation: Price and Enterprise Value Multiples 7.5 Residual Income Valuation 7.6 Private Company Valuation: Income, Market & Asset-Based Approaches 7.7 Choosing and Combining Models: Definitions of Value, Sum-of-the-Parts & Cross-Border Comparability
8 Fixed Income Analysis
8.1 The Term Structure and Interest Rate Dynamics: Spot, Forward & Swap Curves 8.2 The Arbitrage-Free Valuation Framework & Binomial Interest Rate Trees 8.3 Valuation and Analysis of Bonds with Embedded Options: OAS & Duration 8.4 Credit Analysis Models: Structural vs Reduced-Form Approaches 8.5 Credit Default Swaps (CDS): Mechanics, Pricing & Trading Strategies 8.6 Managing Yield-Curve Risk: Key Rate Durations, One-Sided Durations & Yield Volatility Structure
9 Derivatives & Risk Management
9.1 Pricing and Valuation of Forward Commitments: Equity, Fixed Income & FX 9.2 Swap Pricing and Valuation: Interest Rate, Currency & Equity Swaps 9.3 Contingent Claims: Binomial Option Pricing Model & Arbitrage Replication 9.4 The Black-Scholes-Merton (BSM) Model, Option Greeks & Dynamic Hedging 9.5 Interest Rate Options, Swaptions & the Black Model
10 Alternative Investments
10.1 Commodities and Commodity Derivatives: Sectors, Participants, Futures Returns & Indexes 10.2 Private Real Estate Investments: Direct Capitalization, DCF & Cost Approach 10.3 Public Real Estate Investments: REIT Structures, NAV, FFO & AFFO 10.4 Hedge Fund Strategies: Classification, Implementation & Portfolio Role 10.5 Real Estate Investment Features, Due Diligence & Index Construction Biases
11 Portfolio Management
11.1 Using Multifactor Models: Macroeconomic, Fundamental & Statistical Risk 11.2 Measuring and Managing Market Risk: VaR, CVaR & Stress Testing 11.3 Economics & Investment Markets: Business Cycles, Term Structure & Asset Allocation 11.4 Analysis of Active Portfolio Management & The Fundamental Law 11.5 Exchange-Traded Funds (ETFs): Creation, Arbitrage, Tracking & Applications 11.6 Backtesting and Simulation: Procedures, Metrics, Biases & Scenario Analysis
11.2 Measuring and Managing Market Risk: VaR, CVaR & Stress Testing
Key Takeaways Explain This Topic AI
Value at Risk (VaR) quantifies downside tail risk as the minimum expected dollar or percentage loss over horizon T T T at confidence level ( 1 − α ) (1 - \alpha) ( 1 − α ) under normal market conditions.
Parametric VaR assumes normal distributions ( V a R = − [ μ P − z α σ P ] × V P VaR = -[\mu_P - z_\alpha \sigma_P] \times V_P Va R = − [ μ P − z α σ P ] × V P ) and scales with the square root of time ( T \sqrt{T} T ), but underestimates fat tails and non-linear option payoffs.
Historical Simulation VaR revalues current holdings using empirical return histories, while Monte Carlo Simulation generates stochastic paths to price complex optionality and multi-factor distributions.
Conditional VaR (CVaR / Expected Shortfall) measures the expected loss conditional on exceeding VaR ( E [ L ∣ L > V a R ] E[L \mid L > VaR] E [ L ∣ L > Va R ] ) and is subadditive and coherent, complemented by stress testing and reverse stress testing.
Last updated: August 2026
11.2 Measuring and Managing Market Risk: VaR, CVaR & Stress Testing
Core Insight: Market risk is the potential for financial loss resulting from adverse movements in market prices, interest rates, exchange rates, and credit spreads. While traditional volatility ( σ \sigma σ ) measures two-sided dispersion, institutional risk management focuses on downside tail risk . Value at Risk (VaR), Conditional VaR (CVaR), and stress testing form the tripartite framework required to measure, monitor, and control portfolio market risk.
1. Value at Risk (VaR): Formal Definition & Components
Value at Risk (VaR) is the minimum expected loss (in currency terms or as a percentage of portfolio value) over a specified time horizon T T T at a given confidence level ( 1 − α ) (1 - \alpha) ( 1 − α ) under normal market conditions. Pr ( Loss ≥ VaR ) = α ⟺ Pr ( Loss ≤ VaR ) = 1 − α \text{Pr}(\text{Loss} \ge \text{VaR}) = \alpha \iff \text{Pr}(\text{Loss} \le \text{VaR}) = 1 - \alpha Pr ( Loss ≥ VaR ) = α ⟺ Pr ( Loss ≤ VaR ) = 1 − α
The Three Essential Elements of a VaR Statement
Every VaR declaration must specify three parameters:
Time Horizon ( T T T ): The holding period over which risk is measured (e.g., 1 day for trading desks, 10 days for Basel banking regulatory capital, 1 month for pension funds).
Confidence Level ( 1 − α 1 - \alpha 1 − α ): Typically 95% ( α = 5 % \alpha = 5\% α = 5% ) or 99% ( α = 1 % \alpha = 1\% α = 1% ).
Loss Amount: Stated as a percentage ( V a R % VaR_{\%} Va R % ) or currency value ( V a R $ VaR_{\text{\textdollar}} Va R $ ).
Example Interpretation: "A 1-day 95% VaR of $2.5 million means there is a 5% probability (or 1 trading day out of 20) that the portfolio will experience a loss of at least $2.5 million over a single day."
- 2. The Three Primary VaR Estimation Methods VaR Estimation Methodologies
- ├── 1. Parametric (Variance-Covariance) -> Analytical, assumes normal distribution, fast
- ├── 2. Historical Simulation -> Non-parametric, empirical price replays, fat tails
- └── 3. Monte Carlo Simulation -> Stochastic paths, handles complex optionality, flexible
- Method 1: Parametric (Variance-Covariance) VaR
The Parametric VaR method (also known as the analytical or delta-normal method) assumes that asset returns follow a continuous probability distribution—typically the multivariate normal distribution.
Parametric VaR Formula V a R % = − ( μ P − z α σ P ) = z α σ P − μ P VaR_{\%} = -(\mu_P - z_\alpha \sigma_P) = z_\alpha \sigma_P - \mu_P Va R % = − ( μ P − z α σ P ) = z α σ P − μ P V a R $ = V a R % × Portfolio Value = ( z α σ P − μ P ) × V P VaR_{\$} = VaR_{\%} \times \text{Portfolio Value} = (z_\alpha \sigma_P - \mu_P) \times V_P Va R $ = Va R % × Portfolio Value = ( z α σ P − μ P ) × V P
Where:
μ P \mu_P μ P = Mean expected portfolio return over the horizon.
σ P \sigma_P σ P = Standard deviation of portfolio return over the horizon.
z α z_\alpha z α = Critical standard normal z-score corresponding to significance level α \alpha α .
- 95% Confidence (5% tail): z 0.05 = 1.645 z_{0.05} = 1.645 z 0.05 = 1.645 (one-tailed)
- 99% Confidence (1% tail): z 0.01 = 2.326 z_{0.01} = 2.326 z 0.01 = 2.326 (one-tailed)
- 90% Confidence (10% tail): z 0.10 = 1.282 z_{0.10} = 1.282 z 0.10 = 1.282 (one-tailed)
Zero-Mean Assumption: For short time horizons (such as 1 day), the expected daily return μ P \mu_P μ P is negligible and frequently set to zero ( μ P ≈ 0 \mu_P \approx 0 μ P ≈ 0 ), simplifying the equation to V a R = z α σ P V P VaR = z_\alpha \sigma_P V_P Va R = z α σ P V P .
Time Horizon Scaling (Square Root Rule)
Under the assumption of independent and identically distributed (i.i.d.) normal returns, risk scales with the square root of time: σ T = σ 1 × T ⟹ V a R T = V a R 1 × T \sigma_T = \sigma_1 \times \sqrt{T} \implies VaR_T = VaR_1 \times \sqrt{T} σ T = σ 1 × T ⟹ Va R T = Va R 1 × T
Example: If 1-day V a R = $ 1 , 000 , 000 VaR = \text{\textdollar}1,000,000 Va R = $ 1 , 000 , 000 , the 10-day Basel regulatory V a R = $ 1 , 000 , 000 × 10 = $ 3 , 162 , 277 VaR = \text{\textdollar}1,000,000 \times \sqrt{10} = \text{\textdollar}3,162,277 Va R = $ 1 , 000 , 000 × 10 = $ 3 , 162 , 277 .
Method 2: Historical Simulation VaR
The Historical Simulation method applies the actual historical percentage price changes of all underlying risk factors over the past N N N periods (e.g., past 500 or 1,000 trading days) to the current portfolio asset weights .
Step-by-Step Execution:
Collect historical daily return vectors for all portfolio assets over N N N days.
Revalue the current portfolio holdings for each of the N N N historical trading days to generate N N N simulated portfolio returns: R P , t = ∑ w i R i , t R_{P,t} = \sum w_i R_{i,t} R P , t = ∑ w i R i , t .
Sort the N N N simulated portfolio returns from lowest (most negative) to highest.
Identify the cutoff return corresponding to the α \alpha α percentile. For N = 1 , 000 N = 1,000 N = 1 , 000 and a 95% confidence level ( α = 5 % \alpha = 5\% α = 5% ), the VaR is the 50th worst simulated return ( 1 , 000 × 0.05 = 50 1,000 \times 0.05 = 50 1 , 000 × 0.05 = 50 ).
Method 3: Monte Carlo Simulation VaR
Monte Carlo Simulation generates thousands of hypothetical market trajectories based on user-specified stochastic processes, probability distributions, and parameter estimates (drift, volatility, correlations).
Step-by-Step Execution:
Specify stochastic models (e.g., Geometric Brownian Motion, jump-diffusion) and joint distributions for all risk factors.
Draw thousands of correlated random paths (e.g., 10,000 to 100,000 iterations).
Fully reprice all portfolio assets—including non-linear derivatives—at the horizon along each path.
Construct the empirical simulated P&L distribution and identify the ( 1 − α ) (1 - \alpha) ( 1 − α ) percentile loss.
3. Comparative Evaluation of the Three VaR Methods
Evaluation Dimension Parametric (Variance-Covariance) Historical Simulation Monte Carlo Simulation
Distributional Assumption Assumes multivariate normal distribution Non-parametric (no explicit distribution required) Flexible (can specify Student's t, skewed, or jump processes)
Fat Tails & Skewness Fails completely ; severely underestimates tail risk Captures empirically if present in historical sample Captures fully if modeled in stochastic parameters
Non-Linear Derivatives Poor (uses linear delta approximations; ignores gamma/vega) Excellent ; reprices full option contracts directly Excellent ; performs full non-linear revaluation on every path
Computational Complexity Extremely low; instantaneous closed-form matrix math Moderate; requires historical database and daily revaluation Extremely high; requires substantial computing clusters
Primary Weakness / Risk Normality assumption violated in reality; underestimates tail risk Backward-looking; unable to model unprecedented market regimes Severe model risk; vulnerable to mis-specified stochastic assumptions
- 4. Conditional VaR (CVaR / Expected Shortfall)
- The Fundamental Limitation of VaR
Value at Risk is inherently blind to the severity of losses beyond the VaR threshold . For instance, two portfolios may both have a 1-day 99% VaR of $10 million, but Portfolio A's maximum worst-case tail loss is $12 million, whereas Portfolio B (which wrote deep out-of-the-money options) faces a catastrophic tail loss of $100 million. VaR treats both portfolios as having identical risk.
Conditional VaR (CVaR) Definition
Conditional VaR (CVaR), also known as Expected Shortfall (ES) or Expected Tail Loss, measures the expected (average) loss given that the loss exceeds the VaR cutoff : C V a R α = E [ L ∣ L > V a R α ] = 1 α ∫ 0 α V a R u d u CVaR_\alpha = E[L \mid L > VaR_\alpha] = \frac{1}{\alpha} \int_0^\alpha VaR_u \, du C Va R α = E [ L ∣ L > Va R α ] = α 1 ∫ 0 α Va R u d u
In a historical or Monte Carlo simulation with M M M tail observations exceeding VaR, CVaR is simply the arithmetic mean of those M M M tail losses: C V a R α = 1 M ∑ j = 1 M L j for all L j > V a R α CVaR_\alpha = \frac{1}{M} \sum_{j=1}^M L_j \quad \text{for all } L_j > VaR_\alpha C Va R α = M 1 j = 1 ∑ M L j for all L j > Va R α
Coherence and the Subadditivity Property
A risk measure ρ ( ⋅ ) \rho(\cdot) ρ ( ⋅ ) is defined as coherent (Artzner et al.) if it satisfies four mathematical axioms:
Monotonicity: If portfolio X ≤ Y X \le Y X ≤ Y in all states, then ρ ( X ) ≥ ρ ( Y ) \rho(X) \ge \rho(Y) ρ ( X ) ≥ ρ ( Y ) .
Translation Invariance: ρ ( X + c ) = ρ ( X ) − c \rho(X + c) = \rho(X) - c ρ ( X + c ) = ρ ( X ) − c for constant cash c c c .
Positive Homogeneity: ρ ( k X ) = k ρ ( X ) \rho(k X) = k \rho(X) ρ ( k X ) = k ρ ( X ) for any scalar k > 0 k > 0 k > 0 .
Subadditivity: ρ ( X + Y ) ≤ ρ ( X ) + ρ ( Y ) \rho(X + Y) \le \rho(X) + \rho(Y) ρ ( X + Y ) ≤ ρ ( X ) + ρ ( Y ) .
Critical Exam Distinction: VaR is NOT a coherent risk measure because it can violate subadditivity for non-normal, skewed, or fat-tailed distributions (e.g., portfolios with credit default risk or short option positions). Violating subadditivity means that merging two portfolios could result in a combined VaR that is greater than the sum of individual VaRs, falsely implying that diversification increases risk. In contrast, CVaR is mathematically coherent and strictly subadditive under all return distributions.
- 5. Sensitivity Measures, Stress Testing & Risk Limits
- Sensitivity Risk Measures (First- and Second-Order)
While VaR provides a single aggregate loss number, risk managers require granular sensitivity metrics to diagnose specific market risk drivers:
- Asset Class Primary Sensitivity Metric Second-Order / Non-Linear Metric
- Equities Beta ( β \beta β ): Systematic market sensitivity Residual / Idiosyncratic Risk
Fixed Income Modified Duration / Effective Duration: Price sensitivity to parallel yield curve shifts Convexity: Curvature adjustment for large interest rate shifts; Key Rate Duration for curve twists
Options / Derivatives Delta ( Δ \Delta Δ ): Sensitivity to underlying price; Vega ( V \mathcal{V} V ): Sensitivity to implied volatility; Theta ( Θ \Theta Θ ): Sensitivity to time decay Gamma ( Γ \Gamma Γ ): Rate of change of Delta with respect to underlying asset price
Stress Testing and Scenario Analysis
Because VaR relies on statistical distributions calibrated during normal market conditions, it routinely fails during systemic market crises when correlations spike to 1.0 and liquidity vanishes. Stress testing complements VaR by analyzing severe, non-normal tail events:
Historical Scenarios: Replaying historical market shocks across the current portfolio (e.g., the 1987 Black Monday crash, 1998 Russian default/LTCM collapse, 2008 Lehman Brothers bankruptcy, and 2020 COVID-19 liquidity shock).
Hypothetical Scenarios: Modeling plausible forward-looking economic crises that have no historical precedent (e.g., a simultaneous 300 bps stagflation rate spike, 40% crude oil surge, and 200 bps corporate credit spread blowout).
Reverse Stress Testing: An institutional risk practice that starts with the catastrophic outcome —such as a loss that breaches the firm's regulatory capital buffer or causes insolvency—and works backward to identify the specific combination of market shocks that would trigger that failure.
Risk Limits Framework
Institutional capital allocation enforces multiple interlocking risk limits:
Position Limits: Maximum dollar/nominal exposure to any single issuer or asset.
Stop-Loss Limits: Mandatory risk reduction when cumulative losses breach predetermined thresholds.
VaR Limits: Maximum allowable 1-day or 10-day VaR at the desk, division, and firm level.
Scenario / Stress Limits: Maximum permitted loss under designated stress-test scenarios.
- 6. Worked Numerical Example: Parametric VaR, Historical VaR & CVaR
- Part A: Parametric Portfolio VaR Calculation
An investment fund manages a $100 million portfolio split between two asset classes:
Asset 1 (Equities): Weight w 1 = 60 % w_1 = 60\% w 1 = 60% ( V 1 = $ 60 M V_1 = \text{\textdollar}60\text{M} V 1 = $ 60 M ), daily σ 1 = 1.50 % \sigma_1 = 1.50\% σ 1 = 1.50% .
Asset 2 (Bonds): Weight w 2 = 40 % w_2 = 40\% w 2 = 40% ( V 2 = $ 40 M V_2 = \text{\textdollar}40\text{M} V 2 = $ 40 M ), daily σ 2 = 0.80 % \sigma_2 = 0.80\% σ 2 = 0.80% .
Correlation: ρ 12 = 0.20 \rho_{12} = 0.20 ρ 12 = 0.20 . Daily expected returns μ 1 = μ 2 = 0 \mu_1 = \mu_2 = 0 μ 1 = μ 2 = 0 .
Step 1: Calculate Daily Portfolio Volatility ( σ P \sigma_P σ P ) σ P 2 = w 1 2 σ 1 2 + w 2 2 σ 2 2 + 2 w 1 w 2 ρ 12 σ 1 σ 2 \sigma_P^2 = w_1^2 \sigma_1^2 + w_2^2 \sigma_2^2 + 2 w_1 w_2 \rho_{12} \sigma_1 \sigma_2 σ P 2 = w 1 2 σ 1 2 + w 2 2 σ 2 2 + 2 w 1 w 2 ρ 12 σ 1 σ 2 σ P 2 = ( 0.60 ) 2 ( 0.015 ) 2 + ( 0.40 ) 2 ( 0.008 ) 2 + 2 ( 0.60 ) ( 0.40 ) ( 0.20 ) ( 0.015 ) ( 0.008 ) \sigma_P^2 = (0.60)^2 (0.015)^2 + (0.40)^2 (0.008)^2 + 2(0.60)(0.40)(0.20)(0.015)(0.008) σ P 2 = ( 0.60 ) 2 ( 0.015 ) 2 + ( 0.40 ) 2 ( 0.008 ) 2 + 2 ( 0.60 ) ( 0.40 ) ( 0.20 ) ( 0.015 ) ( 0.008 ) σ P 2 = ( 0.36 ) ( 0.000225 ) + ( 0.16 ) ( 0.000064 ) + ( 0.096 ) ( 0.000120 ) \sigma_P^2 = (0.36)(0.000225) + (0.16)(0.000064) + (0.096)(0.000120) σ P 2 = ( 0.36 ) ( 0.000225 ) + ( 0.16 ) ( 0.000064 ) + ( 0.096 ) ( 0.000120 ) σ P 2 = 0.00008100 + 0.00001024 + 0.00001152 = 0.00010276 \sigma_P^2 = 0.00008100 + 0.00001024 + 0.00001152 = 0.00010276 σ P 2 = 0.00008100 + 0.00001024 + 0.00001152 = 0.00010276 σ P = 0.00010276 = 0.010137 ⟹ 1.0137 % per day \sigma_P = \sqrt{0.00010276} = 0.010137 \implies 1.0137\% \text{ per day} σ P = 0.00010276 = 0.010137 ⟹ 1.0137% per day
Step 2: Calculate 1-Day 95% and 99% Parametric VaR
1-Day 95% VaR ( z = 1.645 z = 1.645 z = 1.645 ): V a R 1 d , 95 % = 1.645 × 1.0137 % × $ 100 M = $ 1 , 667 , 537 VaR_{1\text{d}, 95\%} = 1.645 \times 1.0137\% \times \$100\text{M} = \$1,667,537 Va R 1 d , 95% = 1.645 × 1.0137% × $100 M = $1 , 667 , 537
1-Day 99% VaR ( z = 2.326 z = 2.326 z = 2.326 ): V a R 1 d , 99 % = 2.326 × 1.0137 % × $ 100 M = $ 2 , 357 , 856 VaR_{1\text{d}, 99\%} = 2.326 \times 1.0137\% \times \$100\text{M} = \$2,357,856 Va R 1 d , 99% = 2.326 × 1.0137% × $100 M = $2 , 357 , 856
10-Day 99% Regulatory VaR (scaled via 10 \sqrt{10} 10 ): V a R 10 d , 99 % = $ 2 , 357 , 856 × 10 = $ 2 , 357 , 856 × 3.162278 = $ 7 , 456 , 214 VaR_{10\text{d}, 99\%} = \$2,357,856 \times \sqrt{10} = \$2,357,856 \times 3.162278 = \$7,456,214 Va R 10 d , 99% = $2 , 357 , 856 × 10 = $2 , 357 , 856 × 3.162278 = $7 , 456 , 214
Part B: Historical Simulation VaR and CVaR Calculation
A risk manager simulates 500 daily historical P&L returns for a $50 million fund. The 10 worst sorted simulated daily portfolio returns are: Rank 1: − 4.50 % Rank 2: − 3.80 % Rank 3: − 3.20 % Rank 4: − 2.90 % Rank 5: − 2.70 % Rank 6: − 2.50 % Rank 7: − 2.40 % Rank 8: − 2.30 % Rank 9: − 2.20 % Rank 10: − 2.10 % \begin{array}{ccccc}
\text{Rank 1: } -4.50\% & \text{Rank 2: } -3.80\% & \text{Rank 3: } -3.20\% & \text{Rank 4: } -2.90\% & \text{Rank 5: } -2.70\% \\
\text{Rank 6: } -2.50\% & \text{Rank 7: } -2.40\% & \text{Rank 8: } -2.30\% & \text{Rank 9: } -2.20\% & \text{Rank 10: } -2.10\%
\end{array} Rank 1: − 4.50% Rank 6: − 2.50% Rank 2: − 3.80% Rank 7: − 2.40% Rank 3: − 3.20% Rank 8: − 2.30% Rank 4: − 2.90% Rank 9: − 2.20% Rank 5: − 2.70% Rank 10: − 2.10%
Step 1: Compute 1-Day 99% Historical VaR
Significance level α = 1 − 0.99 = 1.0 % \alpha = 1 - 0.99 = 1.0\% α = 1 − 0.99 = 1.0% .
Tail cutoff rank: 500 × 0.01 = 5 th worst observation 500 \times 0.01 = 5\text{th worst observation} 500 × 0.01 = 5 th worst observation .
Rank 5 return is − 2.70 % -2.70\% − 2.70% .
1-Day 99% Historical VaR: 2.70 % × $ 50 M = $ 1 , 350 , 000 2.70\% \times \text{\textdollar}50\text{M} = \mathbf{\text{\textdollar}1,350,000} 2.70% × $ 50 M = $ 1 , 350 , 000 .
Step 2: Compute 1-Day 99% Conditional VaR (CVaR)
CVaR is the average of the losses in the tail beyond the VaR threshold (the 5 worst outcomes, Ranks 1 to 5): Tail Losses = 4.50 % , 3.80 % , 3.20 % , 2.90 % , 2.70 % \text{Tail Losses} = 4.50\%, 3.80\%, 3.20\%, 2.90\%, 2.70\% Tail Losses = 4.50% , 3.80% , 3.20% , 2.90% , 2.70% Mean Tail Loss = 4.50 % + 3.80 % + 3.20 % + 2.90 % + 2.70 % 5 = 17.10 % 5 = 3.42 % \text{Mean Tail Loss} = \frac{4.50\% + 3.80\% + 3.20\% + 2.90\% + 2.70\%}{5} = \frac{17.10\%}{5} = 3.42\% Mean Tail Loss = 5 4.50% + 3.80% + 3.20% + 2.90% + 2.70% = 5 17.10% = 3.42%
1-Day 99% CVaR: 3.42 % × $ 50 M = $ 1 , 710 , 000 3.42\% \times \text{\textdollar}50\text{M} = \mathbf{\text{\textdollar}1,710,000} 3.42% × $ 50 M = $ 1 , 710 , 000 .