Value at Risk (VaR) is the maximum expected loss on a portfolio over a specific time period with a given confidence level — e.g., '5% 1-day VaR of ₹50,000' means there is a 5% chance of losing more than ₹50,000 in one day.
VaR quantifies the worst-case loss with a given probability over a defined horizon. A '95% confidence 1-day VaR of ₹50,000' means: 95% of days, losses will be less than ₹50,000; on 5% of days (roughly 12 trading days per year), losses could exceed ₹50,000. VaR can be calculated using: (1) Historical simulation — use actual historical return distribution; (2) Parametric — assume normal distribution using mean and standard deviation; (3) Monte Carlo — simulate thousands of scenarios. SEBI uses VaR in daily margin requirements for stock market participants.
VaR is a 'worst-case meter' with a confidence level. Think of it as: 'How bad could tomorrow get, and with what probability?' If your ₹10 lakh equity portfolio has a 1-day 95% VaR of ₹30,000 — 95 out of 100 trading days, you'll lose less than ₹30,000. But on those 5 days (about once a month), losses could exceed ₹30,000. VaR is widely used by banks, institutions, and in SEBI's margin framework — less common as a metric in retail mutual fund evaluation.
Define horizon (1 day, 1 month) and confidence level (95% or 99%).
Historical VaR: sort past returns; find the return at the 5th percentile (for 95% confidence).
That percentile return × portfolio value = VaR amount.
SEBI uses VaR for SPAN margin calculations in F&O and for determining debt fund risk.