12b0726c32
* feat: implement robust macro event intelligence system for XAUUSD - Enhanced CSVEventProvider with robust impact and timestamp parsing (handles negative offsets). - Refined category detection for FOMC, interest rate decisions, CPI, and NFP. - Implemented dynamic, linearly decaying risk multiplier for post-event cooldown windows. - Maintained institutional guardrails for major market-moving events. - Verified with 45 comprehensive unit tests. Co-authored-by: saysgrok <279469430+saysgrok@users.noreply.github.com> * feat: implement robust macro event intelligence system for XAUUSD - Enhanced CSVEventProvider with robust impact and timestamp parsing (handles negative offsets). - Refined category detection for FOMC, interest rate decisions, CPI, and NFP. - Implemented dynamic, linearly decaying risk multiplier for post-event cooldown windows. - Maintained institutional guardrails for major market-moving events. - Added docs/macro_intelligence.md. - Resolved python-socketio dependency conflict (downgraded to 4.6.1). - Fixed linting violations in tests. - Verified with 45 comprehensive unit tests. Co-authored-by: saysgrok <279469430+saysgrok@users.noreply.github.com> --------- Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com> Co-authored-by: saysgrok <279469430+saysgrok@users.noreply.github.com>
2.1 KiB
Macroeconomic Event Intelligence System
The macroeconomic event intelligence system provides the XAUUSD trading system with deep awareness of market-moving events. It standardizes data from multiple providers to enforce institutional-grade risk guardrails.
Supported Events
The system specifically targets events with high impact on XAUUSD and USD:
- CPI / Inflation: Consumer Price Index, PCE, PPI, and other inflation indicators.
- NFP / Employment: Non-Farm Payrolls, unemployment rates, and labor market reports.
- FOMC: Federal Reserve meetings, statements, and minutes.
- Interest Rates: Central bank rate decisions and monetary policy statements.
- USD Macro: Broad US economic indicators like GDP, PMI, and Retail Sales.
- Geopolitical: Conflicts, elections, and major geopolitical tensions.
Risk Windows and Guardrails
Lead Windows (Pre-Event)
Before an event occurs, the system monitors a lead window. For major events (CPI, NFP, FOMC, RATES), this window is extended to 120 minutes. During the immediate lead-up (e.g., 60 minutes for major events), execution is strictly blocked to avoid slippage during pre-event volatility.
Digestion Windows (Post-Event)
After an event, the system enforces a digestion window. For major events, this is 180 minutes. Execution is typically blocked for the first 60 minutes of this period.
Dynamic Risk Multiplier
During the digestion window (cooldown), the risk_multiplier linearly recovers from its suppressed value (e.g., 0.25 for major events) back to 1.0. This allows for a disciplined and gradual return to normal position sizing as market volatility subsides.
Data Providers
- MetaAPI: Fetches real-time economic calendar data.
- CSV/JSON: Local file providers for manually curated events or historical analysis.
- TradingView (Mock): Generates synthetic events for testing and pipeline verification.
Implementation Details
The system is implemented in src/data/event_intelligence.py and uses standardized models defined in src/data/event_models.py. It calculates a RiskStatus which is consumed by execution filters to modulate trading activity.