Quantitative Fundamentals: Theory

Quantitative fundamental analysis applies systematic, data-driven methods to financial statement data. Instead of reading annual reports and forming subjective opinions, the quantitative approach converts financial metrics into comparable scores, ranks companies against their peers, and identifies investment candidates through statistical screening.

Why Quantitative?

Traditional fundamental analysis relies on an analyst reading financial statements, building mental models of a company, and making a judgment call. This works well for deep-dive research on a few stocks, but it does not scale. Quantitative fundamentals solve three problems:

  • Scale — Screen hundreds or thousands of stocks simultaneously. A human analyst might cover 20 companies; a quantitative screen covers an entire index in seconds
  • Consistency — The same rules apply to every stock. No cognitive biases, recency effects, or emotional attachment to a particular holding
  • Repeatability — The process can be run weekly, monthly, or quarterly with identical methodology. Results are comparable over time
The Core Idea

Convert raw financial data into normalized scores (typically Z-scores or percentile ranks), then combine them into a composite score that ranks every stock in your universe from best to worst. The top-ranked stocks become your investment candidates.

The Four-Factor Framework

Most quantitative fundamental models organize metrics into factors — groups of related financial characteristics. QuanterLab uses four factors:

Value

Measures how cheaply a stock is priced relative to its financial output. Classic "buy low" metrics. A stock with a low P/E, low P/B, and high earnings yield scores well on Value.

Quality

Measures the financial health and operational efficiency of a company. High ROE, strong margins, low debt, and ample interest coverage indicate a quality business that can sustain its earnings.

Momentum

Measures price performance over various lookback windows. Stocks that have performed well over 3-12 months tend to continue performing well — this is the momentum anomaly, one of the most persistent findings in financial research.

Growth

Measures how rapidly a company is expanding. Revenue growth, earnings growth, and free cash flow growth signal a business that is capturing market share and increasing profitability.

Factor Weighting

No single factor works in all market environments. Value outperforms during recoveries, Momentum during trends, Quality during downturns, and Growth during expansions. The default equal weighting (25% each) provides balanced exposure. Adjusting weights lets you express a market view.

Normalization: Making Metrics Comparable

Raw financial metrics are not comparable across stocks. A P/E of 15 might be cheap for a tech stock but expensive for a utility. Normalization solves this by converting every metric into a relative score within the screening universe.

Z-Score Normalization

Z = (X - μ) / σ Where X = raw value, μ = mean of all stocks, σ = standard deviation

A Z-score of +1.0 means the stock is one standard deviation above the mean for that metric. Higher is better (after inverting metrics where lower is better, like P/E ratio). Z-scores are sensitive to outliers, which is why winsorization (capping extreme values) is applied.

Percentile Rank

An alternative that is more robust to outliers. A percentile rank of 85 means the stock scores better than 85% of the universe on that metric. Less sensitive to extreme values but loses information about the magnitude of differences.

Composite Scoring

The final composite score combines individual metric scores through a two-level weighting system:

  1. Metric-level weights — Within each factor, individual metrics can be set to Off, Low (0.5×), Medium (1×), or High (2×) importance. This lets you emphasize the metrics you trust most
  2. Factor-level weights — The four factors are combined using percentage weights that must sum to 100%. Default is 25% each

The result is a single composite score normalized to a 0-100 scale, where 100 is the best-ranked stock in the universe.

Data Quality

Quantitative screening is only as good as the underlying data. Stocks with missing financial metrics (common for recently listed companies or foreign ADRs) are excluded from scoring. The screener reports which stocks were excluded and why, so you can assess data coverage.

Beyond Screening: The Research Pipeline

Quantitative screening is the first step, not the last. A high composite score identifies candidates — not final investments. The full research pipeline:

Quantitative Research Workflow
  • Screen — Run the multi-factor screener to identify top 10-20 candidates from your target index
  • Validate — Use deep-dive research units (Profitability, Cash Flow, Earnings, Piotroski) to verify the quantitative signal with detailed analysis
  • Assess Risk — Run Monte Carlo simulation and VaR analysis on your shortlist to understand downside potential
  • Optimize — Feed your final selections into portfolio optimization (MVO, HRP, IVO) to determine position sizes
  • Monitor — Track the portfolio in Portfolio Visualizer and re-run the screen periodically to identify new candidates or deteriorating holdings
Back to QuanterLab
Report
Loading report...
Article
Loading article...