CFA Level 3 | Active Equity Investing: Strategies

CFA Level 3 | Portfolio Management Pathway | Learning Module 2

Active Equity Investing: Overview

• Goal: Outperform a passive benchmark.
• Two Broad Approaches:
  • Fundamental (Discretionary): Relies on human judgment, in-depth research of companies, sectors, or markets. Often called “discretionary”.
  • Quantitative (Systematic): Employs rules-based quantitative models and systematic application with limited human discretion. Often called “systematic”.
• Both approaches aim for positive risk-adjusted returns net of fees.
• Some strategies combine elements of both approaches (hybrids).

Fundamental Active Management

• Process:
  
  • Define investment universe and investment thesis (market opportunity and rationale).
  • Prescreen universe using quantitative and/or qualitative criteria.
  • In-depth industry and company analysis, including financial statements.
  • Forecast company performance (earnings, cash flows).
  • Convert forecasts to valuations (intrinsic or relative value) to identify profitable investments.
  • Construct a portfolio with desired risk profile, may involve overweighting/underweighting based on conviction.
  • Rebalance the portfolio with buy and sell disciplines (target prices, stop-loss).
    
• Bottom-Up Strategies: Focus on individual company analysis. Analyze business model, competitive advantages, management.
  
  • Value Investing: Buy stocks at a discount to intrinsic value (low multiples).
    
  • Margin of Safety: Buying earnings and assets inexpensively.
    
  • Relative Value: Compare valuation multiples to peers.
    
  • Contrarian: Invest against market sentiment.
    
  • Deep-Value: Extremely low valuation relative to assets (often distressed).
    
• Growth Investing: Focus on companies with high growth potential (revenues, earnings, cash flow). Higher tolerance for high valuation multiples.
  
  • GARP (Growth at a Reasonable Price): Balances growth and valuation (e.g., low PEG ratio).
    
• Top-Down Strategies: Start with macro analysis (economy, demographics, policies). Allocate across countries, sectors, use derivatives.
  
  • Country/Geographic Allocation: Invest based on regional prospects.
    
  • Sector/Industry Rotation: Allocate based on expected sector returns.
    
  • Volatility-Based: Trade on predictions of market volatility using derivatives.
    
  • Thematic Investing: Identify opportunities based on broad macro, demographic, or technological trends.
    
  • Portfolio Overlays: Use derivatives to adjust macro risk exposures of bottom-up portfolios.
    
• Activist Strategies: Take stakes in companies and advocate for changes to increase value. Focus on strategic, operational, or financial improvements.
  
• Special Situations Strategies: Exploit mispricings from corporate events (divestitures, mergers).
  

Quantitative Active Management

• Process:
  
  • Define investment thesis based on market inefficiencies.
  • Data Acquisition and Processing: Gather and clean large datasets (fundamental, market, survey, unconventional).
    
  • Back-testing: Simulate historical performance of a strategy.
    
  • Information Coefficient (IC): Correlation between factor exposures and subsequent returns (Pearson or Spearman rank). Higher IC indicates higher predictive power.
    
  • Multifactor Model Creation: Select and weight factors to predict returns.
    
  • Strategy Evaluation: Assess model robustness using out-of-sample testing and performance metrics (Sharpe ratio, drawdowns).
    
  • Portfolio Construction: Utilize risk models (variance-covariance matrices) and optimizers.
    
  • Rebalancing: Typically at regular intervals based on model outputs.
• Factor-Based Strategies: Identify and invest in factors (characteristics correlated with returns).
  
  • Rewarded Factors: Historically positive return premium (value, size, momentum, quality).
    
  • Implementation:
    
    • Hedged Portfolios (Long/Short): Go long top quantile, short bottom quantile based on factor.
      
    • Factor-Tilting (Long-Only): Overweight factors in a benchmark-aware portfolio.
    • Factor-Mimicking Portfolios (FMP): Theoretical pure factor exposures.
      
  • Style Factors:
    
    • Value: Low P/E, high book-to-market, etc..
      
    • Momentum: Past winners tend to outperform past losers.
      
    • Growth: High historical or expected growth in fundamentals.
      
    • Quality: High profitability, low leverage, stable earnings (e.g., low accruals).
      
    • Unconventional Factors: Utilize “big data” and alternative data sources.
• Statistical Arbitrage (Stat Arb): Exploit short-term pricing anomalies using statistical and technical analysis (mean reversion, market microstructure).
• Event-Driven Strategies: Profit from market inefficiencies around corporate events (merger arbitrage, restructurings).

Pitfalls in Active Management

• Fundamental Investing:
  
  • Behavioral Biases: Confirmation bias, illusion of control, availability bias.
    
  • Value Trap: Stock appears cheap but has worsening prospects.
    
  • Growth Trap: High growth expectations not met or already priced in.
    
• Quantitative Investing:
  
  • Survivorship Bias: Back-testing with only surviving companies.
    
  • Look-Ahead Bias: Using information not available at the time of investment decision.
    
  • Data Mining/Overfitting: Finding spurious patterns in historical data.
    
  • Transaction Costs: Can significantly reduce net returns.
    
  • Quant Crowding: Many managers following similar strategies, leading to potential large losses during unwinding.

Equity Investment Style Classification

• Divides the investment universe into groups with similar characteristics (value/growth, size, volatility, income).
• Used for manager selection, performance comparison, and asset allocation.
• Approaches:
  
  • Holdings-Based: Aggregates style attributes of individual portfolio holdings (e.g., Morningstar Style Box, MSCI/FTSE Russell indexes).
    
  • Returns-Based: Uses statistical regression to identify style indexes that explain a fund’s historical returns.
    
  • Manager Self-Identification: Based on the fund’s stated investment objective.
• Holdings-based is generally more accurate but requires portfolio holdings data.
• Returns-based can be applied when holdings are not disclosed.
• Significant variation can exist within the same style classification.