📉  Value-at-Risk (VaR)

Value-at-Risk (VaR)#

A threshold loss unlikely to be exceeded at a given confidence over a horizon.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What it is#

Value-at-Risk summarizes downside risk in a single number: the maximum loss over a holding period \(h\) that will not be exceeded with confidence \(\alpha\) (typically 95% or 99%) — anything worse occurs only with probability \(1 - \alpha\). Formally it is the \(\alpha\)-quantile of the loss distribution (the negative \(\alpha\)-quantile of returns):

\[\mathrm{VaR}_\alpha = -F_r^{-1}(\alpha), \qquad \Pr\!\left(L > \mathrm{VaR}_\alpha\right) = 1 - \alpha,\]

with \(F_r\) the return CDF and \(L\) the loss.

Where it comes from#

VaR was introduced by J. P. Morgan’s RiskMetrics (1994) and enshrined by the Basel framework for bank regulatory capital. It is estimated by historical simulation (the empirical quantile over a rolling window), parametric methods (assume a normal / t distribution and scale by volatility, often via GARCH), or Monte Carlo.

Its blind spot#

VaR says nothing about how bad losses beyond the threshold are, and it is not coherent — it can violate subadditivity, so a diversified portfolio’s VaR may exceed the sum of its parts. Expected Shortfall (CVaR) — the average loss given VaR is breached — repairs both and is coherent.


Theme: Risk & Probabilistic Forecasting  ·  All terminology



See also

Source article Adapted (context, re-expressed) in our own words from: Value-at-Risk (VaR) (insightful-data-lab.com).

Tags: purpose: reference topic: terminology level: advanced