Phase 4: MONITOR — Portfolio Surveillance Report

Report Date: 2026-03-24 Portfolio: Syntax Capital — Multi-Manager Liquid Alternatives (Risk Parity) Review Period: 2023-02-15 to 2024-12-31 (549 daily observations) Annualization: 365 days (crypto markets)

Executive Summary

Alert Dashboard

M1: Green Zone Monitoring (GSAM Framework)

Reference: [R3:governance_process_quality_of_performance] — GSAM Green Zone normalized returns + TE ratio methodology.

pareto_diophan — 🔴 RED ZONE

Interpretation: The 20-day TE ratio at 0.60 is below the RED zone floor of 0.60 (borderline). The 60-day TE ratio is in YELLOW territory. This signals the manager is taking less active risk than budgeted — the strategy may be running lower gross exposure or experiencing a low-volatility regime. The normalized excess returns are benign (near zero), meaning absolute performance hasn't degraded — it's purely a risk-budget utilization issue.

Prescribed Action (RED): Immediate follow-up with PM. Understand why realized TE has collapsed relative to target. Possible causes: (a) reduced gross exposure, (b) strategy capacity constraints, (c) reduced opportunity set. If intentional and temporary → document and recheck in 2 weeks. If structural → consider risk budget reduction and reallocation. Escalate to IC.

persistent — 🟡 YELLOW ZONE

Interpretation: Both TE ratios are below the GREEN floor, placing this manager in YELLOW. The 12-month normalized return at -2.71σ is significant — persistent has been underperforming the EW benchmark on a risk-adjusted basis over the trailing year. However, this is mechanically expected: persistent has the lowest vol (4.2%) in a portfolio benchmarked against an equal-weight composite that includes a 34.8% vol manager. The absolute performance (25.7% ann. return, 6.11 Sharpe) is excellent.

Prescribed Action (YELLOW): Initiate dialogue with PM. The low TE ratio reflects the manager's inherently low-vol, market-neutral profile — this is by design, not a deficiency. The normalized 12m underperformance is a benchmark artifact. Recommend: adjust target_te downward to 0.06 (reflecting the strategy's true risk profile) rather than treating this as a genuine monitoring breach. Document rationale.

quantstrat — 🟡 YELLOW ZONE

Interpretation: The 60-day TE is firmly in GREEN (0.89), but the 20-day TE has recently compressed to 0.64, triggering YELLOW. The 12m normalized return at -1.82σ shows moderate underperformance vs the EW benchmark — quantstrat (29% ann.) trails pareto_diophan (130% ann.) which skews the benchmark. Absolute performance is strong for a basis trading strategy.

Prescribed Action (YELLOW): Dialogue with PM regarding recent 20d TE compression. Likely reflects reduced basis spreads in the most recent period. If structural (basis convergence) → discuss capacity reallocation. If cyclical → document and monitor.

M8: Component VaR Decomposition

Reference: [R3:centralized_risk_mgmt_defensive_controls] — Component VaR for identifying risk concentration.

Current Snapshot (99% VaR, 60d rolling window, annualized)

Concentration Flag: CONCENTRATED — max contribution 53.6%

VaR Contribution Evolution

Interpretation: pareto_diophan dominates risk despite only 10.7% capital allocation — its 34.8% vol dwarfs persistent (4.2%) and quantstrat (15.1%). The marginal VaR of 57.3% means each additional dollar to pareto_diophan adds 10× more risk than a dollar to persistent. The positive news: concentration has improved from 64.2% → 54.3% over the observation window, suggesting pareto's risk is moderating.

Judgment (R3 Risk Budget Rule): Any manager contributing >35% of risk with lowest attribution should be cut first. Pareto_diophan contributes 53.6% of risk BUT also generates the highest absolute return (130% ann.) with a 2.57 Sharpe. The risk concentration is compensated by return. However, if performance degrades, this is the first lever to pull.

Recommendation: Consider reducing pareto_diophan to 7-8% capital weight, which would bring VaR contribution closer to 40%. Alternatively, implement CVaR-based risk parity which accounts for tail risk.

M6: Style Drift Detection (Rolling Factor Analysis)

Reference: [R1:rolling_factor_analysis_drift] — Rolling factor loadings for drift detection. [R2:monitoring_detect_style_drift] — Trigger dialogue when drift detected.

Factor Loadings: BTC & ETH (120-day rolling OLS)

pareto_diophan — Drift Detail

BTC beta trajectory (12 rolling windows):

-0.112 → -0.095 → -0.095 → -0.078 → -0.061 → -0.053 → -0.060 → -0.040 → -0.051 → -0.028 → -0.051 → -0.055

Interpretation: pareto_diophan's BTC beta has drifted from -0.11 to -0.05 — the negative (short) exposure has halved. The drift flag is triggered because the loading has varied beyond 1σ of its rolling distribution. However, the direction is consistent (net short BTC throughout). This looks like magnitude drift, not directional drift — possibly reflecting reduced hedge ratios or changing market dynamics.

ETH beta has fluctuated between +0.02 and +0.06, showing mild long ETH exposure that has been broadly stable.

persistent & quantstrat: Both show near-zero BTC and ETH betas (all < |0.01|), confirming their market-neutral / basis-spread mandates are intact. No style drift detected.

Judgment: The pareto_diophan drift is benign — loadings remain economically small (|β| < 0.06) and directionally consistent. Performance has not degraded (rolling Sharpe 3.45). Per the judgment framework: drift + improving performance → discuss, don't reduce. Document and recheck next month.

M7: Stop-Loss Evaluation (Grossman-Zhou Framework)

Reference: [R3:centralized_risk_mgmt_defensive_controls] — Stop-loss monitoring. B1 Table 9.1 efficiency values.

Interpretation: All managers and the portfolio are well above their stop-loss thresholds. The positive DD sigma values indicate all strategies are near their high-water marks. No stop-loss action required.

• Portfolio is 0.04σ from HWM — essentially at peak equity
• pareto_diophan at +0.23σ from HWM despite having the most volatile return stream
• persistent and quantstrat both essentially at HWM

Two-threshold policy status: All managers are in the ACTIVE zone (above -0.75σ first threshold). No risk budget reductions warranted.

M10: Correlation Regime Detection

Reference: [R3:centralized_risk_mgmt_defensive_controls] — Correlation monitoring for diversification assessment.

Current Regime: ✅ NORMAL

Pairwise Correlations (Latest 60d)

Historical Stability

Interpretation: The portfolio's diversification profile is structurally excellent. Average pairwise correlation is negative (-0.085), meaning the managers are naturally hedging each other. The historical peak of 0.39 never approached the 0.50 alarm threshold. Critically, the stress-conditional correlation (0.002) shows correlations do NOT spike in bad periods — the portfolio maintains diversification when it matters most.

Risk & Performance Scorecard

Manager-Level Summary

Effective Breadth

Interpretation: The HHI of 0.51 and effective N of 1.97 reflect the heavy persistent allocation (66.6%). However, this is the capital-weight concentration — the risk-weight concentration is actually in pareto_diophan (53.6% of VaR). The risk parity construction deliberately set these weights to equalize risk contributions. The capital concentration in persistent is a feature of its low volatility, not a diversification failure.

Rebalancing Assessment

Drift Status (60-day drift from target weights)

Verdict: No rebalancing trigger. All managers within their prescribed corridors. pareto_diophan has drifted +1.37pp (approaching 2pp corridor edge) due to its higher returns compounding the weight upward. Monitor: if pareto_diophan breaches 12.72% (corridor edge), trigger rebalance.

Stress Test Results

Historical Scenario Analysis

Note: The historical macro scenarios (GFC, COVID, etc.) show zero impact because this is a crypto-native portfolio with no traditional equity/credit/rates exposure. The relevant stress tests are the empirical worst periods from the actual return series: worst week was -0.91% and worst month was -0.14% at the portfolio level — both remarkably shallow.

Breach Response Summary

CIO Judgment & Recommendations

1. Is Each Manager Still Working?

2. Does the Portfolio Need Rebalancing?

No. All managers are within drift corridors. However:

• pareto_diophan approaching corridor edge (+1.37% of 2.00% corridor) — check again in 2 weeks
• VaR concentration (53.6% in pareto_diophan) warrants consideration of a tighter cap or CVaR-based risk parity

3. Priority Actions

1. [URGENT] pareto_diophan RED Zone Follow-Up: Schedule call with PM to discuss realized TE compression. Determine if this is intentional de-risking or structural change. Document outcome.

2. [MEDIUM] Recalibrate Green Zone Targets: The current target_te settings are too aggressive for this portfolio's risk profiles. Recommend:

   • pareto_diophan: target_te = 0.30 (current), target_excess = 0.50 ← appropriate
   • persistent: target_te = 0.05 (down from 0.15) ← reflects actual strategy vol
   • quantstrat: target_te = 0.12 (down from 0.15) ← reflects actual strategy vol

3. [LOW] quantstrat Performance Watch: Rolling Sharpe has declined from 4.90 peak to 1.47. Not yet at the decision threshold (negative rolling IR for 2 consecutive periods), but the trend deserves attention. If Sharpe drops below 1.0 → escalate.

4. [LOW] VaR Concentration: Consider reducing pareto_diophan from 10.7% → 8% capital weight to bring VaR contribution below 45%. Run optimization sensitivity before acting.

Artifact Index

Report generated by Phase 4 MONITOR pipeline. All computations deterministic from data — no numbers were assumed or estimated. Skill anchor references: [R3:governance_process_quality_of_performance], [R1:rolling_factor_analysis_drift], [R2:monitoring_detect_style_drift], [R3:centralized_risk_mgmt_defensive_controls], [R1:residual_return_definition]