Home / Use cases

Valcision use cases

Capital-at-risk analysis

Connect scenario losses across risk groups, examine portfolio tail risk, and explore how diversification and allocation assumptions shape the result.

See how risks combine across the portfolio.

Standalone exposures do not tell the whole story. Capital-at-risk analysis brings risk components into a common scenario framework so a team can examine aggregate losses and the contribution of individual groups.

  • How large are portfolio losses in the adverse part of the scenario distribution?
  • How do dependency assumptions change aggregate risk?
  • What diversification benefit appears when risks are analyzed jointly?
  • How should the selected risk measure be attributed to products or risk groups?

Define what the numbers represent.

Scope, units, and horizon

Identify the products and risk groups in scope, choose a reporting currency, and use a consistent time horizon. Define the loss measure before combining outputs from different models.

Scenarios and dependencies

Supply aligned monetary loss values for each component in each scenario. Make distribution and dependency assumptions explicit when generating joint outcomes.

Use a consistent sign convention: positive losses and negative gains in the financial risk workflow. Every component must be represented in every scenario; absent exposures require explicit zero losses rather than missing records.

Build a connected view of capital at risk.

  1. Develop component loss estimates. Prepare data and models for the risks being studied. Convert results to compatible monetary losses with a common horizon and currency.
  2. Represent joint outcomes. Use the supported simulation workflow to explore distributions and dependencies, or bring externally generated, aligned scenario losses into the analysis.
  3. Aggregate by scenario. Combine component losses within each scenario before calculating portfolio risk. Preserve the scenario identity through the workflow.
  4. Measure the adverse outcomes. Review value at risk and expected shortfall at the selected confidence level. Compare portfolio measures with standalone risk estimates.
  5. Explore allocation and sensitivity. Apply the chosen allocation methodology and compare how results change under alternative dependencies, distributions, or scenario assumptions.

Repeat the analysis as assumptions change, keeping the model inputs and calculation choices visible alongside the outputs.

Illustrative use case: aggregate and allocate portfolio tail risk.

All figures below are fictional and deliberately simplified. They illustrate a workflow and calculation; they are not customer results, fitted model outputs, or forecasts.

A small scenario set for a transparent calculation

A risk team brings monetary loss scenarios for credit, interest-rate, and operational risk. The following five equally weighted scenarios use one currency and one horizon; amounts are in millions of dollars.

The example uses an 80% confidence level so the empirical quantile and tail average can be followed directly. Five scenarios are a teaching device, not a sufficient dataset for a practical tail-risk estimate.

1. Align component losses by scenario

ScenarioCreditRateOperationalTotal
1$1,000,000$5,000,000$0$6,000,000
2$2,000,000$4,000,000$0$6,000,000
3$3,000,000$3,000,000$0$6,000,000
4$4,000,000$2,000,000$1,000,000$7,000,000
5$5,000,000$1,000,000$2,000,000$8,000,000

All values are losses. The credit and rate losses move in opposite directions in this deliberately constructed set. That relationship is an assumption of the example, not an empirical claim about those risk groups.

2. Calculate portfolio VaR and expected shortfall

The sorted portfolio losses are 6, 6, 6, 7, and 8. Using the financial workflow’s empirical quantile convention, the 80% VaR is the observation at rank ceil(0.80 × 5) = 4.

Follow the portfolio calculation

Portfolio VaR at 80% = $7 million.

Expected shortfall at 80% = $8 million. The worst 20% of this equally weighted set consists of scenario 5, so the tail average is its $8 million total loss.

VaR identifies the selected quantile; expected shortfall averages the selected adverse tail. They describe different parts of the same loss distribution.

3. Compare standalone and aggregate risk

Using the same empirical convention, standalone 80% VaRs are $4 million for credit, $4 million for rate risk, and $1 million for operational risk. Their sum is $9 million.

Portfolio VaR is $7 million, giving a $2 million difference relative to the sum of standalone VaRs. This apparent diversification benefit depends on the joint scenario structure.

To test that dependence, pair the credit losses 1–5 with rate losses 1–5 and operational losses 0, 0, 0, 1, 2. The marginal loss sets remain the same, but portfolio losses become 2, 4, 6, 9, 12. Portfolio 80% VaR rises to $9 million and expected shortfall rises to $12 million.

4. Attribute expected shortfall to risk groups

In the original scenario set, the 80% expected-shortfall tail contains only scenario 5. The component contributions are therefore $5 million for credit, $1 million for rate risk, and $2 million for operational risk. They sum to the portfolio expected shortfall of $8 million.

This is an expected-shortfall attribution, not an allocation of the $7 million VaR. A percentile-based allocation workflow uses its own selected methodology and convention.

5. Keep the calculation connected in Valcision

The team connects aligned component losses to scenario aggregation, risk measurement, and reporting. It compares joint assumptions and keeps the selected confidence level, quantile convention, loss definitions, and allocation method alongside the outputs.

What the example establishes

Portfolio risk must be calculated from joint scenario totals, rather than inferred by adding standalone quantiles. The example also shows how dependency assumptions can change tail risk even when component loss distributions are unchanged.

These results illustrate analytical capital-at-risk assessment. They are not regulatory capital requirements or a recommendation for a capital buffer.

Keep the measure and its allocation connected.

OutputWhat it helps explain
Portfolio loss distributionThe range of combined losses across the supplied or simulated scenarios.
Value at riskThe loss quantile at the chosen confidence level.
Expected shortfallThe average loss in the selected adverse tail.
Standalone comparisonHow joint portfolio risk differs from separately measured components.
Risk allocationHow the chosen methodology attributes a portfolio risk measure to its components.

Valcision provides scenario-based financial risk measures and a separate simulation-based allocation workflow. Select the methodology to match the measure: expected-shortfall contributions and percentile-based allocation approaches answer different questions.

Common questions

Is this a regulatory capital calculation?

The documented scenario measures support analytical capital-at-risk assessment. They do not, by themselves, calculate regulatory capital or establish regulatory compliance.

Can I add the standalone VaRs to get portfolio VaR?

Portfolio VaR is calculated from aggregate scenario losses. Summing standalone VaRs generally gives a different quantity because component outcomes and their dependencies affect the portfolio distribution.

Can allocation results be negative?

Some supported percentile-based allocation methods retain negative marginal contributions when adding a component reduces the portfolio percentile. Interpret the result with the method, sign convention, and scenario set in view.

What should be reviewed before relying on the result?

Review loss definitions, scenario completeness, dependency assumptions, simulation sample size, confidence level, and sensitivity to alternative assumptions. More simulation draws alone do not establish that tail estimates are accurate.

Bring your analytical need into view.

Share your data, modeling question, and the decisions your team needs to support.

Discuss your use case