Adding $1,000,000 of unmodeled value raises starting assets to $11,000,000 and reduces value coverage to approximately 90.909%. Covered P/L remains -$1,471,875. Neither a full-portfolio return nor a full ending value is estimated. This is hypothetical, not a client or provider result.
Preserve the inventory before calculating coverage
Begin with the four default holdings: $4 million of equities, $3 million of Treasuries, $2 million of corporate bonds and $1 million of USD cash. Add a fifth holding, Private asset, with a starting USD value of $1 million and type unmodeled. The resulting inventory totals $11 million. Nothing has been sold, reweighted or replaced, and the supported holdings retain their original sensitivities.
Use the same equity-led preset: equities -30%, government yields +50 basis points and credit spreads +150 basis points. Its currency shocks do not affect the USD-denominated supported holdings. Selecting unmodeled for the private asset is a statement about this calculator's scope, not a claim that private investments have no market exposure. No public-equity beta or bond duration is invented to fill the gap.
Keep unknown outcomes distinct from zero
For the unsupported holding, the engine returns no local return, FX return, USD return, P/L or ending value. Those fields are null, not measured zeros. Its starting value remains in the inventory total. The aggregate P/L sums only the supported results, so the known loss stays -$1,471,875, exactly as in the four-holding example. The private asset contributes no estimated P/L, which is different from an estimate of no loss.
Covered starting value remains $10 million, giving coverage of $10 million / $11 million * 100 = 90.9090909%, or 90.909% rounded. The uncovered share is approximately 9.091%. These are market-value proportions, not confidence levels, probabilities or estimates of the fraction of total risk captured. A small unsupported position can still have important leverage, liquidity or contractual exposures.
Use the right denominator for each question
Dividing the covered loss by covered starting value gives -14.71875%. That is the modeled-sleeve return on $10 million. Dividing the same loss by the full $11 million inventory gives approximately -13.380682%. That second number is the modeled contribution to total starting value, not a complete portfolio return. Its smaller magnitude reflects a larger denominator, not a demonstrated diversification benefit from the private asset.
The supported holdings end at $8,528,125 in aggregate. This is covered ending value only. Adding the private asset's unchanged starting mark would produce $9,528,125, but doing so would introduce an unsupported zero-return assumption. The engine therefore withholds the full ending value. The full return must also remain not estimated until a defensible scenario outcome for the missing holding is incorporated.
| Measure | Calculation or treatment | Reported result |
|---|---|---|
| Total starting value | $10,000,000 + $1,000,000 | $11,000,000 |
| Covered starting value | Four supported holdings | $10,000,000 |
| Unmodeled starting value | Private asset retained in inventory | $1,000,000 |
| Value coverage | $10,000,000 / $11,000,000 * 100 | 90.909% (rounded) |
| Covered P/L | Unchanged supported-holding sum | -$1,471,875 |
| Modeled-sleeve return | -$1,471,875 / $10,000,000 * 100 | -14.71875% |
| Modeled contribution / total | -$1,471,875 / $11,000,000 * 100 | -13.380682% (rounded) |
| Covered ending value | $10,000,000 - $1,471,875 | $8,528,125 |
| Full ending value | Private-asset outcome unavailable | Not estimated |
| Full-portfolio return | Private-asset outcome unavailable | Not estimated |
A starting valuation is not a stress valuation
The FCA's private-market valuation review explains that infrequent trading and limited price discovery require judgment in private-asset valuations. A recent carrying value does not by itself supply sensitivity to this scenario. For this example, the $1 million mark establishes only the starting denominator. It does not establish what a buyer would pay after the assumed equity, rate and credit moves.
A separate analysis could examine cash flows, financing terms, comparable valuations and a documented valuation range. It would need consistent dates and scenario assumptions before its output could be combined with the supported sleeve. Do not quietly replace the missing result with a stale appraisal or assume an immediate sale at that mark. Any overlay should remain distinguishable from the engine's original reference output.
Make the coverage limitation travel with the result
The Basel Committee's stress-testing principles call for users to understand risk coverage and model limitations. Here that means displaying the $1 million gap alongside the loss, including in summaries. Even the covered sleeve uses simplified fixed sensitivities and excludes costs, income, default and forced liquidation. A successfully reconciled subtotal is useful evidence of arithmetic consistency, but it cannot validate the missing valuation or establish complete economic coverage.
Questions for resolving the missing exposure
Who owns the private-asset valuation, and how old is its starting mark? What evidence could support a scenario-sensitive estimate? Are capital calls, borrowing or redemption restrictions relevant to funding needs? Would a decision change across a credible valuation range? Preserve the incomplete report while those questions remain open. Do not label the lower contribution percentage a better portfolio outcome merely because the unknown asset enlarges the denominator.
Review checkpoints
- Keep the $1 million unsupported asset in the $11 million starting inventory.
- Coverage is 90.909%; covered P/L remains -$1,471,875.
- The -13.380682% modeled contribution is not the full-portfolio return.
- Withhold full ending value and full return until the gap is resolved.
Source notes
Continue the analysis
Coverage and Missing Positions in a Stress Test
Measure what a stress calculation includes, expose unsupported holdings, and interpret partial losses without silently treating missing risk as zero.
Mapping Portfolio Holdings to Risk Factors
Create an auditable bridge from securities and contracts to signed factor exposures, consistent units, and reconciled scenario results.
Read the stress report before accepting its headline
Audit scope, coverage, sign conventions, denominators, loss bridges, and limitations before interpreting a portfolio stress total.
Keep stress-report revisions explainable and reproducible
Assign ownership, preserve report versions, bridge changed results, and prevent corrections or assumption changes from rewriting the record.
First edition, 16 October 2026. Hypothetical values are teaching inputs, not observed portfolio or provider results.