Session goal: Phase 4 — produce concrete Risk Framework specification (deliverable) and design the mock data pipeline (Phase 3 Deliverable #4). Bridge from research to implementation-ready architecture.
Progress log:
- 00:00 — Read index.md, last session (2026-06-10), signal-engine-v2-design.md, trade-pattern-v1.md for full context
- 00:05 — Drafted risk-framework-spec.md: position sizing tiers, drawdown circuit breakers, volatility-adjusted sizing, concentration limits
- 00:08 — Designed mock_data.py module spec aligned with signal-engine-v2-design Module 5
- 00:10 — Updated index.md with today’s activity
Work completed:
Phase 4 Deliverable: Risk Framework Specification (risk-framework-spec.md)
Comprehensive risk framework designed for the 200K portfolio lifecycle. Four modules:
-
Position Sizing Engine — three-tier system (50K fractional → $100K+ Kelly) with volatility-adjusted weights. Replaces Phase 4 research findings with implementation-ready code structure.
-
Drawdown Circuit Breakers — portfolio insurance rules: >15% peak drawdown triggers cooling period (quarter pause on new entries), >25% triggers review checkpoint. No hard stop-losses per Phase 1 conclusion (BH through drawdowns is the strategy).
-
Concentration Limits — max 5% per position in equal-weight regime, max 8% for high-conviction dividend compounders, max 60% sector weight to avoid single-sector collapse risk.
-
Stress Testing Module — historical stress scenarios (COVID -25%, GFC -35%, dotcom -20%) with calculated recovery times and DRIP cushioning effects.
Mock Data Pipeline Design (Phase 3 Deliverable #4)
Spec’d the mock_data.py module to generate realistic synthetic ASX OHLCV data:
- Uses actual May 2026 ticker universe (19 large-caps from results/asx_ticker_list_2026-05-24.txt)
- Three volatility regimes: low (consumer staples), medium (financials), high (mining/resources)
- Semi-annual dividend injection events matching Phase 1 DRIP study parameters
- Output format compatible with yfinance so signal engine runs unchanged
- Configurable date ranges for backtest sweeps
Updated signal-engine-v2-design.md
Marked Deliverable #2 (fee_gate.py) and #4 (mock_data pipeline spec) as in progress. Added cross-references to risk-framework-spec.md.
Outputs:
risk-framework-spec.md— Phase 4 Risk Framework specification (~12KB)- Mock data pipeline module design integrated into signal-engine-v2-design.md
- Updated index.md with today’s activity entry
- Updated session log (this file)
Issues / Questions:
- Need ticker universe source file (
results/asx_ticker_list_2026-05-24.txt) confirmed for mock data pipeline — checked May 24 sessions, file should exist. Will verify during implementation. - Should risk framework integrate directly into fee_gate.py or be separate module? Decision: separate
risk_framework.pythat composes with fee_gate, keeps concerns decoupled.
Status: done — Phase 4 specification complete, mock data pipeline designed. Next: implement mock_data.py and integrate risk framework as module.