Methodology

The Three Layers

Clavix analyzes your portfolio across three interconnected layers:

Macro - Broad market conditions: interest rates, inflation, GDP growth, geopolitical events, Federal Reserve policy. These factors set the overall risk tone for all portfolios.

Sector - Industry and sector-level movements: earnings surprises, regulatory changes, supply chain shifts, sector rotation. These affect groups of related positions.

Position - Company-specific news: earnings misses, leadership changes, product recalls, legal issues. These directly affect individual holdings.

Risk Scoring

Each position receives a risk score derived from the intersection of all three layers. The score reflects the magnitude and probability of downside risk relative to your current position size and portfolio concentration.

Scores are expressed as letter grades (A through F) and signal the severity of current risk:

Signal Actions

Every risk score maps to one clear action - designed to cut through noise and tell you exactly what to do:

Daily Digest

Each morning, Clavix delivers a personalized briefing organized in three parts:

Market Context - Key macro events from the prior 24 hours and what they mean for markets broadly.

Where You're Exposed - How macro and sector moves connect to your specific holdings, ranked by risk severity.

Clear Actions - A ranked list of positions with action recommendations and the reasoning behind each one.

Data Sources

Clavix aggregates data from multiple sources:

• Brokerage API connections (read-only) for holdings data

• Licensed financial news and data feeds for market event detection

• Public market data providers for price, volume, and fundamentals

• Alternative data sources (e.g., options flow, sentiment indices) for early signal detection

Model Limitations

No model is perfect. Clavix's risk analysis has known limitations:

• Lag - News events take time to propagate into scores. Real-time market moves may not be reflected immediately.

• Black swan events - Sudden, unpredictable market shocks (natural disasters, geopolitical crises) may overwhelm historical patterns the model relies on.

• Correlation assumptions - The model assumes certain securities behave similarly to their historical patterns. Unusual market regimes can break these assumptions.

• Sentiment limits - News sentiment analysis may miss sarcasm, spin, or delayed information that affects market prices.