A statistical measure estimating the maximum expected loss of a portfolio over a given time period at a specified confidence level.
Value at Risk (VaR) is a statistical risk measure that quantifies the maximum expected loss a portfolio could suffer over a defined time horizon at a given confidence level. A daily VaR of INR 5 lakh at 95% confidence means there is a 95% probability that the portfolio will not lose more than INR 5 lakh in a single day — equivalently, there is a 5% chance the loss will exceed this amount.
There are three primary methods for calculating VaR. The parametric (variance-covariance) method assumes returns are normally distributed and uses the portfolio's standard deviation. The historical simulation method uses actual past returns without distributional assumptions. The Monte Carlo method generates thousands of random scenarios based on statistical parameters. Each has trade-offs between simplicity, accuracy, and computational cost.
In Indian markets, VaR is used extensively by exchanges, clearing corporations, and brokers for margin calculation. NSE Clearing uses a VaR-based margin system (SPAN) for derivative positions, calculating the margin required to cover the expected maximum one-day loss at 99% confidence. This directly affects how much capital traders need to maintain in their margin accounts.
For portfolio managers and PMS providers, VaR is a standard risk reporting metric. SEBI requires disclosure of risk measures in PMS and AIF (Alternative Investment Fund) communications. A well-diversified equity portfolio of large-cap Indian stocks might show a daily VaR of 2-3% at 95% confidence, while a concentrated mid-cap portfolio could show 5-6% or more.
VaR has important limitations. It tells you the minimum loss at the confidence threshold, not the maximum. The 5% of days that exceed VaR could involve losses far greater than the VaR figure — this "tail risk" is captured by Expected Shortfall (Conditional VaR), which averages the losses beyond the VaR threshold. After the 2008 financial crisis, regulators globally supplemented VaR with stress testing, recognising that VaR underestimates risk during market dislocations.
Formula
Parametric VaR = Portfolio Value × Z-score × σ × √t (where σ = standard deviation, t = time horizon)India Context
NSE uses VaR-based SPAN margin system for F&O. SEBI requires risk disclosure for PMS and AIF. Clearing corporations compute VaR daily for margin requirements.