Fixed Income Validation

1. Calculation Name

Fixed-Income Data Validation Suite

2. Description and Mathematical Formula

The validation plugin inspects inbound positions before calculations. Typical checks include:

• Required field coverage: \( \mathrm{Coverage} = \frac{\text{non-null rows}}{\text{total rows}} \)
• Tolerance checks: \( |x_{\text{portfolio}} - x_{\text{benchmark}}| \leq \tau \)
• Duration consistency: \( |\mathrm{DV01}_{\text{sum}} - \mathrm{Reported\ DV01}| \leq \epsilon \)
• FX exposure balance: \( \sum \mathrm{FX\ Notional} \approx 0 \) for hedged portfolios

3. Input Sample Data

Check ID	Description	Sample Metric	Threshold	Status	Notes
FI-V-001	Mandatory columns present (ISIN, MV, Yield)	100% coverage	100%	Pass	No missing cells
FI-V-004	Sum of DV01 vs reported total	144.7 vs 145.0	±0.5	Pass	Difference = 0.3
FI-V-007	FX hedge residual	0.8mm USD	±1.0mm	Pass	Within tolerance
FI-V-010	Negative amortization flag	2 rows	0	Warn	Review expected amort schedule
FI-V-014	Benchmark mapping coverage	97%	99%	Fail	3 securities unmapped

4. Mathematical Solution

• Coverage example: \( \frac{980 \text{ valid rows}}{980} = 100\% \) (Check FI-V-001).
• DV01 reconciliation: \( |144.7 - 145.0| = 0.3 \leq 0.5 \) → Pass.
• FX residual: \( |0.8| \leq 1.0 \) million → Pass.
• Amortization: 2 violations > 0 threshold → Warning.
• Benchmark mapping: \( 97\% < 99\% \) → Fail, highlighting missing benchmark IDs.

5. Sample Python and R Code

Python

results = [
    {"check": "FI-V-001", "metric": 1.00, "threshold": 1.00},
    {"check": "FI-V-004", "metric": abs(144.7 - 145.0), "threshold": 0.5},
    {"check": "FI-V-007", "metric": abs(0.8), "threshold": 1.0},
    {"check": "FI-V-010", "metric": 2, "threshold": 0},
    {"check": "FI-V-014", "metric": 0.97, "threshold": 0.99},
]

for row in results:
    status = "Pass"
    if row["metric"] > row["threshold"]:
        status = "Warn" if row["check"] == "FI-V-010" else "Fail"
    print(row["check"], status)

R

results <- tibble::tibble(
  check = c("FI-V-001", "FI-V-004", "FI-V-007", "FI-V-010", "FI-V-014"),
  metric = c(1.00, abs(144.7 - 145.0), abs(0.8), 2, 0.97),
  threshold = c(1.00, 0.5, 1.0, 0, 0.99)
)

results <- results %>%
  mutate(
    status = case_when(
      check == "FI-V-010" & metric > threshold ~ "Warn",
      metric > threshold ~ "Fail",
      TRUE ~ "Pass"
    )
  )
results

6. Output Table

Severity	Checks	Action
Pass	FI-V-001, FI-V-004, FI-V-007	No action required
Warn	FI-V-010	Review amortization schedule
Fail	FI-V-014	Map missing benchmark IDs before running attribution

7. Conclusion

Validation guardrails prevent calculation surprises. Documenting the tests—and their thresholds—alongside the outputs helps operations teams close data gaps before running sensitive fixed-income analytics in FinFacts.