Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Finance›Value at Risk (VaR)
Regression model

Value at Risk (VaR)

Value at Risk (Historical, Parametric, Monte Carlo) · Also known as: VaR, value-at-risk, delta-normal VaR, historical simulation VaR, Riske Maruz Değer (VaR — Historical, Parametric, MC)

Value at Risk is a financial risk measure that estimates the maximum loss a position or portfolio could suffer over a fixed holding period at a given confidence level. It is the standard benchmark in risk management and regulatory capital calculations, developed in the textbook tradition of Jorion (2007) and the Basel market-risk framework.

ScholarGate
  1. Regression model
  2. v1
  3. 2 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Value at Risk
ARIMAConditional Value-at-RiskGARCHMONTE-CARLO-SIMULATIONRealized VolatilityCopula ModelsDCC-GARCHExtreme Value TheoryImportance SamplingRisk Parity Portfolio

When to use it

Use VaR when you hold a continuous-valued financial position over time and need a single, interpretable summary of downside risk for risk management or capital requirements. It needs a reasonable history of returns — at least about 100 observations, and ideally 250 or more — and the series should be stationary. The parametric form is quick but assumes normality and so understates tail risk; historical and Monte Carlo forms relax that. VaR is unreliable on short data histories and on non-stationary series, where differencing or a simpler volatility measure is preferable.

Strengths & limitations

Strengths
  • Produces a single, intuitive number — the threshold loss at a stated confidence level and horizon — that managers and regulators readily understand.
  • Offers three complementary estimation routes (historical, parametric delta-normal, Monte Carlo) so the assumptions can be matched to the data.
  • Is the established standard in market-risk management and Basel regulatory capital calculations.
Limitations
  • VaR is not sub-additive: the VaR of a combined portfolio can exceed the sum of the parts, so it is not a coherent risk measure — CVaR (expected shortfall) is the coherent alternative.
  • It says nothing about the size of losses beyond the threshold, hiding tail severity.
  • The parametric (normal) form systematically understates tail risk; estimates are unreliable on short histories (n < 250) and non-stationary series.

Frequently asked

What is the difference between historical, parametric, and Monte Carlo VaR?

Historical simulation reads the loss quantile directly from past returns and makes no distributional assumption. Parametric (delta-normal) VaR assumes returns are normal and uses the mean and standard deviation, which is fast but understates fat tails. Monte Carlo simulates many return paths from a model and takes the tail quantile, flexibly handling complex positions at higher computational cost.

Why is VaR not a coherent risk measure?

A coherent measure must be sub-additive — diversifying should never increase measured risk. VaR can violate this: the VaR of a combined portfolio can exceed the sum of the individual VaRs. Conditional Value at Risk (CVaR / expected shortfall) is sub-additive and therefore coherent.

How much data do I need to estimate VaR?

At least about 100 return observations, and ideally 250 or more. On shorter histories the tail quantile is unstable and a simpler realized-volatility measure is preferable.

What is backtesting and why does it matter?

Backtesting checks how often realised losses actually breach the estimated VaR over time. If breaches occur far more or far less often than the confidence level implies, the model is miscalibrated. Backtesting is what validates a VaR model before it is trusted.

Sources

  1. Jorion, P. (2007). Value at Risk: The New Benchmark for Managing Financial Risk (3rd ed.). McGraw-Hill. ISBN: 978-0071464956
  2. Basel Committee on Banking Supervision (2019). Minimum Capital Requirements for Market Risk. Bank for International Settlements. link ↗

How to cite this page

ScholarGate. (2026, June 1). Value at Risk (Historical, Parametric, Monte Carlo). ScholarGate. https://scholargate.app/en/finance/value-at-risk

Related methods

ARIMAConditional Value-at-RiskGARCHMONTE-CARLO-SIMULATIONRealized Volatility

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • ARIMAEconometrics↔ compare
  • Conditional Value-at-RiskFinance↔ compare
  • GARCHEconometrics↔ compare
  • MONTE-CARLO-SIMULATIONDecision-making↔ compare
  • Realized VolatilityFinance↔ compare
Compare side by side →

Referenced by

Conditional Value-at-RiskCopula ModelsDCC-GARCHExtreme Value TheoryImportance SamplingRisk Parity Portfolio

Similar methods

Conditional Value-at-RiskTail Risk MeasuresVaR BacktestingExtreme Value TheoryMean-Variance Portfolio OptimizationGARCH ModelRealized VolatilityVariance Reduction for Monte Carlo

Related reference concepts

Risk ManagementFinancial EconometricsCopula ModelsMonte Carlo MethodsStatistical Decision TheoryMaximum Likelihood Estimation

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Value at Risk (Value at Risk (Historical, Parametric, Monte Carlo)). Retrieved 2026-07-21 from https://scholargate.app/en/finance/value-at-risk · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Jorion (textbook benchmark); popularised by RiskMetrics / J.P. Morgan
Year
2007
Type
Financial risk measure
Estimator
Lower-tail quantile of the return distribution (historical, parametric delta-normal, or Monte Carlo)
Outcome
continuous (loss)
MinSample
100
Coherent
No — VaR is not sub-additive; CVaR is coherent
Related methods
ARIMAConditional Value-at-RiskGARCHMONTE-CARLO-SIMULATIONRealized Volatility
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account