Repository navigation
v1.2.0 - Docker 1-Click, Dynamic Slippage, Institutional Risk Metrics & Out-of-Sample Validation π
π Quant Backtest Pro v1.2.0 β Institutional Realism, 1-Click Docker & Out-of-Sample Validation
We are proud to release Quant Backtest Pro v1.2.0! This milestone introduces institutional quantitative analytics, market execution realism, out-of-sample forward testing, and complete containerization.
π Highlights in v1.2.0
π³ 1-Click Docker Containerization & CI Test Gate
- Multi-Stage Dockerfile: High-performance Node.js 20 Alpine production build.
- docker-compose.yml: 1-command launch with SQLite volume persistence (
docker compose up -d). - Automated CI Test Gate: 101 automated test cases running on every commit via GitHub Actions.
- Official ROADMAP.md: Strategic milestone documentation from v1.0 to v2.0.
π Institutional Realism & Quant Risk Metrics
- Sortino Ratio: Downside deviation risk-adjusted return measurement.
- Calmar Ratio: Annualized return relative to Maximum Drawdown.
- System Quality Number (SQN - Van Tharp): Mathematical system reliability rating (Holy Grail π, Excellent β, Good π, Average, Poor
β οΈ ). - Dynamic Slippage Model: Realistic market order slippage simulation.
- Overnight Swap Rates: Accurate financing fee deductions across
00:00 UTC.
π‘οΈ Out-of-Sample Walk-Forward Validation (Anti-Overfitting)
- Train/Test Split (70% IS / 30% OOS) in SL/TP Grid Optimizer.
- Efficiency Index & Robustness Rating (Robust β, Moderate, Overfitted
β οΈ ) badges.
πΎ Dataset Management & Cleaned CSV Export
- Multi-Batch Online Crawler: Paginated fetching up to 50,000 candles with date range filters.
- SQLite Dataset Library: 1-click loading and pre-seeded institutional datasets.
- Download CSV: 1-click export of sanitized historical price data.
-
Time-Travel Replay Bar: Interactive calendar date-time picker with
$O(\log N)$ Binary Search.
π§ͺ Verification:
- 101 / 101 Automated Tests Passing (100%)
- TypeScript 0 errors, Vite production build verified