Value at Risk methods 2026: VaR Exam Guide (IIBF)
Among all the quantitative tools a risk manager wields, Value at Risk methods remain the single most examined family of techniques in the IIBF Risk Management certificate, and for good reason: VaR distils a portfolio's downside into one number that boards, regulators, and traders can all act on. This article unpacks the three classical approaches — historical simulation, the variance-covariance (parametric) method, and Monte Carlo simulation — explains their assumptions and pitfalls, and shows how VaR connects to backtesting, Expected Shortfall, and 2026 regulatory expectations. Master these Value at Risk methods and you will handle the market-risk section of the paper with genuine confidence rather than rote memory.
At its core, VaR answers one question: over a given horizon and at a chosen confidence level, what is the maximum loss the portfolio is unlikely to exceed? A "1-day 99% VaR of Rs 5 crore" means there is only a 1% chance of losing more than Rs 5 crore tomorrow.
Historical Simulation Method
The historical simulation approach is the most intuitive of the Value at Risk methods. It takes actual historical returns of the risk factors — say the last 250 to 500 trading days — revalues the current portfolio under each of those past scenarios, sorts the resulting profit-and-loss outcomes, and reads off the loss at the chosen percentile. No distributional assumption is needed; the data speaks for itself.
Its strengths and weaknesses are exam favourites:
- Pros — no assumption of normality, captures fat tails and real correlations present in the data, easy to explain to non-specialists.
- Cons — assumes the future resembles the chosen past window, reacts slowly to sudden regime changes, and gives equal weight to old and recent observations unless a weighting scheme is applied.
Because it is transparent and model-light, historical simulation is widely used by Indian banks for their trading books. To see how VaR feeds capital, connect it with the market-risk capital concepts in the CAIIB risk syllabus and drill the numbers on our practice platform.
Variance-Covariance (Parametric) Method
The variance-covariance method — also called the parametric or analytical method — assumes portfolio returns are normally distributed. It estimates the mean and standard deviation of returns and derives VaR directly using a z-score: for 95% confidence the multiplier is 1.65, and for 99% it is 2.33. VaR is then computed as the portfolio value multiplied by the z-score multiplied by the return volatility over the chosen horizon.
- Pros — computationally light, fast, and elegant for large linear portfolios; requires only volatilities and correlations.
- Cons — the normality assumption understates tail risk (real markets have fat tails), and it handles options and other non-linear instruments poorly because it ignores convexity (gamma) effects.
This method is prized for speed but dangerous when portfolios contain significant optionality, which is precisely why examiners test the normality caveat. Sharpen your recall of the 1.65 and 2.33 multipliers and the underlying assumptions with quick drills on our risk concept-match game before the exam.
Monte Carlo Simulation Method
Monte Carlo simulation is the most flexible and computationally intensive of the Value at Risk methods. It specifies a statistical process for the risk factors, generates thousands of random simulated scenarios, fully revalues the portfolio under each, and builds an empirical loss distribution from which VaR is read at the chosen percentile.
- Pros — handles non-linear instruments (options, structured products) accurately, can incorporate any distribution including fat-tailed ones, and models complex path-dependent payoffs.
- Cons — computationally heavy, dependent on the quality of the assumed model and inputs (model risk), and can create false precision if the underlying assumptions are wrong.
Monte Carlo is the tool of choice for portfolios rich in derivatives, where the linear approximations of the parametric method break down. Because model risk is central here, it links naturally to a bank's broader model-validation and governance framework, which we cover in more depth on the IIBF risk blog. Test your grasp of all three methods side by side with a themed set on our Risk Management mock tests.
Backtesting, Expected Shortfall and 2026 Regulation
Computing VaR is only half the discipline; validating it is the other half. Backtesting compares predicted VaR against actual daily P&L and counts "exceptions" (days where losses exceeded VaR). Basel's traffic-light approach classifies models as green, amber, or red based on the exception count over 250 days, with amber and red triggering higher capital multipliers.
Two further concepts define the 2026 landscape:
- Expected Shortfall (ES) — under Basel's Fundamental Review of the Trading Book (FRTB), the internal-models approach shifts from 99% VaR to 97.5% Expected Shortfall, which measures the average loss beyond the threshold and better captures tail risk. Indian banks are aligning with FRTB timelines set by RBI.
- Stressed calibration — models must be calibrated to periods of significant financial stress, ensuring VaR/ES does not lull risk managers during calm markets.
For the authoritative global standard, consult the Bank for International Settlements, whose Basel Committee publishes the market-risk rules, and track RBI's domestic adoption via our regulatory updates hub.
Two limitations of VaR deserve special attention because examiners return to them. First, VaR is not sub-additive in general — the VaR of a combined portfolio can occasionally exceed the sum of the individual VaRs, which violates the intuition that diversification should reduce risk. Expected Shortfall does not suffer this flaw, being a coherent risk measure, and this is a key theoretical reason regulators favour ES under FRTB. Second, VaR says nothing about the magnitude of losses in the tail: a 99% VaR is silent on whether the 1% of bad days lose slightly more than the threshold or catastrophically more. This "tail blindness" is precisely why stress testing and scenario analysis complement, rather than replace, VaR in a mature risk framework. A well-prepared candidate can state the horizon-scaling rule too: to convert a 1-day VaR to a 10-day VaR under the square-root-of-time approximation, multiply by the square root of 10, roughly 3.16 — a favourite numerical question.
Frequently Asked Questions
Related study material
Go deeper with the full chapter notes and the complete article hub for this subject:
What are the three main Value at Risk methods?
The three classical VaR methods are historical simulation (using actual past returns), the variance-covariance or parametric method (assuming normally distributed returns), and Monte Carlo simulation (generating thousands of random scenarios). Each differs in assumptions, accuracy for non-linear instruments, and computational cost.
Why does the parametric method understate tail risk?
The variance-covariance method assumes returns follow a normal distribution, but real market returns have fatter tails, meaning extreme losses occur more often than the normal curve predicts. It also handles options poorly because it ignores convexity, so it can understate true downside risk.
What is backtesting in VaR?
Backtesting compares each day's predicted VaR against the actual profit or loss and counts exceptions where losses exceeded VaR. Basel's traffic-light framework rates models green, amber, or red over a 250-day window, with poorer results attracting higher capital multipliers.
How does Expected Shortfall differ from VaR?
VaR gives the loss threshold at a confidence level, but says nothing about how bad losses are beyond it. Expected Shortfall (ES) measures the average loss in the tail beyond that threshold. Under Basel FRTB, the internal-models approach uses 97.5% ES instead of 99% VaR to better capture tail risk.
Conclusion: Turn VaR Theory Into Marks
Value at Risk methods reward candidates who can contrast the three approaches, recall the key multipliers, and link VaR to backtesting and Expected Shortfall under FRTB. Study them as a connected toolkit rather than isolated formulas and the market-risk paper becomes far more predictable. Ready to prove it? Attempt a full-length market-risk set on our Risk Management mock tests and cement your preparation today.
Practice this topic
Take a free mock test, download chapter PDFs, or watch a video class — all included on iibf.store.
