Skip to content

Configuration

Learn how to customize the analysis framework by creating and modifying configuration files.

Configuration Structure

All analysis parameters are defined in YAML configuration files stored in the configs/ directory. Each configuration defines:

  • Stock universe - Which stocks to analyze
  • Screening criteria - Quality, value, growth, and risk thresholds
  • Valuation models - DCF, RIM parameters
  • Output options - Results format and limits

Basic Configuration

Here's a simple configuration example:

name: "my_analysis"
description: "Custom stock screening strategy"

universe:
  region: "US"
  min_market_cap: 1000  # $1B minimum

quality:
  min_roic: 0.12        # 12% minimum ROIC
  min_roe: 0.15         # 15% minimum ROE
  max_debt_equity: 0.6   # Maximum 60% debt/equity

value:
  max_pe: 25            # Maximum P/E ratio
  max_pb: 3.5           # Maximum P/B ratio

growth:
  min_revenue_growth: 0.03  # Minimum 3% revenue growth

risk:
  max_beta: 1.5         # Not too volatile

max_results: 50
sort_by: "composite_score"

Configuration Sections

Universe Selection

Define which stocks to analyze:

universe:
  region: "US"                    # US, EU, JP, ALL
  min_market_cap: 1000           # $1B minimum (in millions)
  max_market_cap: 100000         # $100B maximum (optional)

  # Sector filters
  sectors: ["Technology", "Healthcare"]           # Include only these
  exclude_sectors: ["Real Estate", "Utilities"]  # Exclude these

  # Custom stock list
  custom_tickers: ["AAPL", "GOOGL", "MSFT"]

  # Predefined universes
  pre_screening_universe: "sp500"  # Use S&P 500 stocks
  top_n_by_market_cap: 100        # Top 100 by market cap

Quality Criteria

Financial strength and stability metrics:

quality:
  min_roic: 0.12                 # Return on Invested Capital
  min_roe: 0.15                  # Return on Equity  
  min_current_ratio: 1.2         # Current assets / Current liabilities
  max_debt_equity: 0.6           # Total debt / Total equity
  min_interest_coverage: 5.0     # EBIT / Interest expense

Value Criteria

Valuation and price metrics:

value:
  max_pe: 25                     # Price / Earnings ratio
  max_pb: 3.5                    # Price / Book ratio
  max_ev_ebitda: 15              # Enterprise Value / EBITDA
  max_ev_ebit: 20                # Enterprise Value / EBIT
  max_p_fcf: 30                  # Price / Free Cash Flow

Growth Criteria

Business growth and expansion metrics:

growth:
  min_revenue_growth: 0.03       # Minimum revenue growth rate
  min_earnings_growth: 0.05      # Minimum earnings growth rate
  min_fcf_growth: 0.02           # Minimum free cash flow growth
  min_book_value_growth: 0.03    # Minimum book value growth

Risk Assessment

Risk and volatility parameters:

risk:
  max_beta: 1.5                  # Maximum market risk (beta)
  min_liquidity_ratio: 1.0       # Minimum liquidity measures
  max_concentration_risk: 0.3    # Geographic/customer concentration
  cyclical_adjustment: true      # Apply sector-specific adjustments

Valuation Models

Configure valuation model parameters:

valuation:
  models: ["dcf", "rim"]         # Models to run
  scenarios: ["bear", "base", "bull"]  # Scenario analysis

  # DCF specific parameters
  dcf_years: 10                  # Projection period
  terminal_growth_rate: 0.025    # Terminal growth rate (2.5%)
  risk_free_rate: 0.04           # Override risk-free rate

  # RIM specific parameters
  rim_years: 10                  # Projection period
  required_return: 0.10          # Required return rate

Output Options

Control results and formatting:

max_results: 50                  # Maximum stocks in final results
sort_by: "composite_score"       # Sort criterion
generate_reports: true           # Generate detailed reports
save_data: true                  # Save intermediate data

Pre-built Configurations

The framework includes several ready-to-use configurations:

Conservative Value Strategy

# configs/conservative_value.yaml
quality:
  min_roic: 0.15      # High quality requirement
  min_roe: 0.18
  max_debt_equity: 0.4  # Low debt

value:
  max_pe: 20          # Reasonable valuation
  max_pb: 2.5

Aggressive Growth Strategy

# configs/aggressive_growth.yaml
growth:
  min_revenue_growth: 0.15  # High growth requirement
  min_earnings_growth: 0.20

value:
  max_pe: 40          # Accept higher multiples
  max_pb: 8.0

S&P 500 Full Analysis

# configs/sp500_full.yaml
universe:
  pre_screening_universe: "sp500"

max_results: 503      # Include all S&P 500 stocks

Creating Custom Configurations

1. Copy Existing Configuration

Start with a similar strategy: 

cp configs/sp500_full.yaml configs/my_strategy.yaml

2. Modify Parameters

Edit the YAML file to match your investment criteria:

name: "my_dividend_strategy"
description: "Focus on dividend-paying quality companies"

quality:
  min_roe: 0.12
  max_debt_equity: 0.8

value:
  max_pe: 30

# Add dividend-specific criteria
custom_filters:
  min_dividend_yield: 0.02  # 2% minimum dividend yield

3. Test Configuration

poetry run python scripts/systematic_analysis.py configs/my_strategy.yaml --save-csv

Advanced Configuration

Sector-Specific Adjustments

risk:
  cyclical_adjustment: true

# Automatic adjustments by sector:
# - Technology: Higher P/E acceptable, higher growth expected
# - Utilities: Lower growth acceptable, stable margins expected
# - Energy: Higher volatility acceptable, cyclical patterns considered

Multiple Scenarios

valuation:
  scenarios: ["bear", "base", "bull"]

# Generates valuations under different assumptions:
# - Bear: Conservative growth, higher discount rates
# - Base: Expected growth, market discount rates  
# - Bull: Optimistic growth, lower discount rates

Configuration Validation

The framework validates all configurations and will show helpful error messages:

# Invalid configuration example
ERROR: quality.min_roe must be between 0.0 and 1.0, got 15.0
HINT: Use decimal format (0.15 for 15%)

Best Practices

Start Conservative

  • Begin with stricter criteria
  • Gradually relax constraints if needed
  • Document your reasoning for each threshold

Sector Considerations

  • Technology: Higher P/E ratios acceptable (25-40)
  • Utilities: Lower growth requirements (3-5%)
  • Banks: Different debt metrics (use book value ratios)

Backtesting

  • Test configurations on historical data
  • Verify results align with investment thesis
  • Adjust based on performance analysis

Example Workflows

Value Investing Focus

quality:
  min_roic: 0.12
  min_roe: 0.15
  max_debt_equity: 0.5

value:
  max_pe: 18
  max_pb: 2.0
  max_ev_ebitda: 12

Growth Stock Screening

growth:
  min_revenue_growth: 0.10
  min_earnings_growth: 0.15

value:
  max_pe: 45  # Accept higher multiples for growth
  max_pb: 8.0

Dividend Aristocrats

quality:
  min_roe: 0.10
  max_debt_equity: 0.8

# Focus on stability over growth
growth:
  min_revenue_growth: 0.00  # Accept flat growth

# Custom dividend requirements would be added here

Next Steps