Skip to content

Systematic Investment Analysis Framework

A configuration-driven, objective approach to investment analysis that eliminates conversational bias and provides consistent, reproducible results with comprehensive backtesting capabilities.

Framework Overview License Poetry CI

Quick Start

# Install dependencies
poetry install

# Launch interactive dashboard (recommended)
poetry run python scripts/dashboard_server.py
# Opens automatically at http://localhost:8080

# Or run command-line analysis
poetry run python scripts/systematic_analysis.py configs/sp500_full.yaml --save-csv

# View results
cat sp500_full_screen_*_results.csv

Key Features

🎯 AI-Controlled Dual Analysis Approach

Systematic Screening: AI models (like Claude) run automated analysis pipelines on large stock universes, eliminating human bias in the filtering process.

AI Deep-Dive Analysis: AI models then use specialized tools to perform qualitative analysis on companies that pass the systematic filters.

⚙️ Configuration-Driven

Define your investment criteria in YAML files. No code changes needed to adjust screening parameters.

📊 Comprehensive Analysis

  • Quality Assessment: ROIC, ROE, debt levels, liquidity ratios
  • Value Analysis: P/E, P/B, EV/EBITDA ratios vs. thresholds
  • Growth Evaluation: Revenue/earnings growth, sustainability
  • Risk Assessment: Financial, market, and business risk factors
  • Valuation Models: DCF and RIM models with multiple scenarios

🏢 Sector Context

Automatically adjusts expectations based on sector characteristics: - Technology: Higher growth, higher multiples expected - Utilities: Lower growth, stable margins expected
- Energy: High cyclicality, volatile margins expected

📈 Global Market Coverage

  • Screens entire S&P 500 universe and international markets
  • Japanese markets (TOPIX, Berkshire holdings) - Warren Buffett's recent focus
  • European markets (FTSE, DAX) and other international opportunities
  • 50+ financial metrics evaluated with currency-aware analysis
  • Multiple valuation models with international considerations

🤖 AI-Controlled Research Process

  • AI models execute the entire research workflow autonomously
  • AI runs systematic screening, interprets results, and performs deep-dive analysis
  • AI uses specialized tools for data gathering, analysis, and report generation
  • Seamless integration with Claude Desktop and Gemini ecosystems

Philosophy

This framework empowers AI models to conduct comprehensive investment research:

  • AI-Driven Systematic Screening: AI models run objective, bias-free filtering of large stock universes
  • AI-Controlled Deep Analysis: AI models autonomously use tools for qualitative analysis of promising candidates
  • Human-Configurable: Humans define investment criteria in YAML files, AI executes the research
  • AI-Reproducible: AI follows consistent methodology for systematic screening
  • End-to-End AI Workflow: Quality → Value → Growth → Risk → Valuation → AI Deep-Dive pipeline

Output Formats

The framework generates:

  1. Executive Summary - High-level results and top picks
  2. Detailed Stock Reports - Comprehensive analysis for each stock
  3. CSV Export - Data for further analysis with pass/fail indicators
  4. JSON Export - Structured data for integration

Why AI-Controlled Investment Research?

Traditional investment research often suffers from:

  • Human bias - Cherry-picking supportive data, emotional decisions
  • Inconsistency - Different analysis methods for different stocks
  • Scale limitations - Humans can't systematically analyze hundreds of stocks
  • Time constraints - Manual deep analysis is extremely time-consuming

This AI-controlled framework provides:

  • AI-Executed Systematic Filtering - AI objectively screens large universes (500+ stocks) using consistent methodology
  • AI-Driven Deep Analysis - AI performs conversational analysis of promising candidates (10-50 stocks) using specialized tools
  • Hybrid AI Approach - Quantitative screening rigor + AI qualitative insights
  • AI-Scalable Process - AI handles entire markets while maintaining analysis depth autonomously

Perfect for empowering AI models to provide comprehensive investment research with both breadth and depth.

Getting Started

  1. Installation - Set up the environment
  2. Quick Start - Run your first analysis
  3. Configuration - Customize screening criteria

AI-Controlled Workflow

Step 1: AI Runs Systematic Screening

# AI model executes systematic analysis on US markets
poetry run python scripts/systematic_analysis.py configs/sp500_full.yaml --save-csv

# AI model analyzes international markets (Warren Buffett's Japanese favorites)
poetry run python scripts/systematic_analysis.py configs/japan_buffett_favorites.yaml --save-csv

# AI automatically filters to 25-50 companies globally that pass all criteria

Step 2: AI Performs Deep-Dive Analysis

# AI model uses specialized investment research tools to analyze promising candidates
# AI automatically accesses:
# - Claude Desktop investment tools
# - Gemini AI research tools  
# - Custom analysis workflows
# - All available data sources and research capabilities

Step 3: AI Generates Investment Recommendations

AI combines quantitative screening results with qualitative insights to provide comprehensive investment analysis and recommendations.

Learn More