Value at Risk VaR
A statistical estimate of the worst loss expected over a set period at a chosen confidence level, under normal conditions.
What it is
Value at Risk (VaR) is a single number summarizing downside risk: the maximum loss a position or portfolio is expected to suffer over a fixed time horizon at a stated confidence level. A 1-day 95% VaR of $1M means there is roughly a 5% chance of losing more than $1M in a single day (and a 95% chance the loss stays at or below that). It is always tied to both a horizon (1 day, 10 days) and a confidence level (95%, 99%), so quoting a VaR number without those two parameters is meaningless.
Why it matters
VaR became the standard risk dashboard number for banks, funds, and regulators because it compresses messy return distributions into one comparable figure. The critical pitfall: VaR says nothing about how bad losses get beyond the cutoff — it ignores the size of tail events, so two portfolios with identical VaR can have wildly different worst-case behavior. It also tends to understate risk in crises (returns are fatter-tailed than the normal-distribution assumption), and ordinary VaR is not "subadditive," meaning a combined portfolio's VaR can exceed the sum of its parts, perversely penalizing diversification. Conditional VaR (Expected Shortfall) was created to fix the tail-blindness.
How it's calculated
Pick a horizon and confidence level, then estimate the loss distribution by one of three methods. Historical simulation re-prices the portfolio over past returns and reads off the relevant percentile; the parametric (variance-covariance) method assumes a normal distribution and scales the standard deviation by the matching z-score (about 1.65 for 95%, 2.33 for 99%); Monte Carlo simulates thousands of random scenarios and takes the percentile of simulated losses. VaR is the loss at that percentile of the distribution.
How Quintarthai uses it
Use VaR alongside the volatility and drawdown figures on a company's deep-analysis page to size how much a position could move against you, and read it together with maximum-drawdown and the Sharpe ratio. The Knowledge Base covers the complementary tail-risk and risk-adjusted-return measures it should be paired with.