BQuant - Quantitative Research Toolkit

BQuant is a universal toolkit for quantitative research of financial markets. The project starts with MACD zone analysis as the first use case, but the architecture is designed for exploring various aspects: technical indicators, chart patterns, candlestick formations, time series, and machine learning applications.

🔧 Key Features

Zone Analysis - Universal Pipeline v2.1

• Universal API: Works with any indicator (MACD, RSI, AO, custom) through fluent builder pattern
• 5 Detection Strategies: zero_crossing, threshold, line_crossing, preloaded, combined
• Advanced Analysis: Swing, divergence, volume, volatility strategies with automatic feature extraction
• Statistical Testing: Automatic hypothesis tests and clustering analysis
• Caching Support: Performance optimization with memory and disk caching

Core Features

• Universal Configuration System: Flexible settings for data sources, indicators, and analysis
• Extensible Indicator Library: Includes optimized built-in indicators and supports external libraries like pandas-ta and TA-Lib
• ML Readiness: A modular structure prepared for future machine learning integration
• Rich Visualization Tools: Create interactive financial charts and statistical plots with Plotly and Matplotlib
• Performance-Oriented: Features a two-level caching system and performance monitoring tools
• Command-Line Interface: Provides a simple CLI for quick analysis and data management

🚀 Quick Start

Installation

# Install in development mode
pip install -e .

# Install with optional dependencies
pip install -e .[dev,notebooks]

Basic Usage - Universal Pipeline v2.1

from bquant.data.samples import get_sample_data
from bquant.analysis.zones import analyze_zones

# Load sample data
data = get_sample_data('tv_xauusd_1h')

# Universal Pipeline - работает с любым индикатором
result = (
    analyze_zones(data)
    .with_indicator('pandas_ta', 'rsi', length=14)
    .detect_zones('threshold', indicator_col='RSI_14', 
                  upper_threshold=70, lower_threshold=30)
    .analyze(clustering=True)
    .build()
)

print(f"Found {len(result.zones)} zones")
print(f"Statistics: {result.statistics}")

MACD-зоны в одну строку

from bquant.analysis.zones import analyze_macd_zones

result = analyze_macd_zones(data)  # пресет поверх того же пайплайна

MACDZoneAnalyzer (и обёртки create_macd_analyzer / analyze_macd_zones из bquant.indicators.macd) удалён. Замена — пресет выше либо полный билдер analyze_zones(...). Сам индикатор MACD не менялся.

pandas-ta indicators in one line

from bquant.indicators import LibraryManager

# Load external libraries (pandas-ta, TA-Lib when installed)
LibraryManager.load_all_libraries()

# "Simple way" to access any pandas-ta indicator discovered dynamically
rsi = LibraryManager.create_indicator('pandas_ta', 'rsi', length=14)
result = rsi.calculate(data)
print(result.data.tail())

See the LibraryManager documentation for more examples.

Command Line

# List available sample datasets
bquant list

# Detect and analyze zones (MACD by default) on a sample dataset
bquant analyze tv_xauusd_1h

# Any oscillator: the zone vocabulary follows the indicator
bquant analyze tv_xauusd_1h --indicator rsi

# Structured output, for programs rather than for reading
bquant analyze --json --no-chart

# Analyze and save the chart to a file
bquant analyze mt_xauusd_m15 -o chart.html

See the CLI guide for every flag and the JSON schema.

📋 Project Structure

This is a monorepo that contains:

• bquant/ - Python package (for PyPI)
• research/ - Jupyter notebooks and experiments
• scripts/ - Automation scripts
• data/ - Data storage
• tests/ - Test suite
• docs/ - Documentation

🛠️ Development

Setting up development environment

# Create virtual environment
python -m venv .venv

# Activate (Windows)
.venv\Scripts\activate

# Activate (Linux/Mac)
source .venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Install in development mode with all extras
pip install -e .[full]

Running tests

pytest tests/ -v

📚 Documentation

Universal Pipeline v2.1

• Quick Start - 5 минут до первого результата
• API Reference - полная документация Universal Pipeline
• Examples - готовые примеры для всех индикаторов
• Migration Guide - переход с deprecated API

Complete Documentation

• API Documentation - Справочник API
• Tutorials - Обучающие материалы
• Examples - Примеры использования
• Developer Guide - Руководство разработчика

Architecture

• Two-Layer Design: Simplification from 3 to 2 layers
• Zero Hardcode: ZERO hardcoded indicators, full universality
• Design Patterns: Strategy, Dependency Injection, Builder, Registry
• Test Suite: 1155 tests, green on every release

🎯 Roadmap

• Phase 1 (Completed): Core functionality (data loading, processing, validation), advanced MACD analysis, and statistical engine.
• Phase 2 (In Progress): Extended visualization options, implementation of Time Series and other indicator analysis modules (currently stubs).
• Phase 3 (Planned): Full machine learning integration, chart pattern recognition, and enhanced automation pipelines.

📄 License

MIT License - see LICENSE file for details.

🤝 Contributing

Contributions are welcome! Please read our contributing guidelines and submit pull requests.

📞 Contact

• Author: kogriv
• Email: kogriv@gmail.com
• Repository: https://github.com/kogriv/bquant