Documentation | Giraffe Research

About Giraffe Research

We are a market intelligence and research platform built for retail investors. By applying rigorous statistical methods to current fundamental and market data, we deliver systematically researched insights that are highly accessible, consumable, and designed to complement standard online financial information.

Data Sourcing & Integrity

Understanding our pipelines, refresh frequencies, and calculations.

Data Sourcing & Quality

We ingest data from premier commercial providers and open-source libraries via systematic pipelines. To ensure institutional-grade integrity, we run rigorous internal validation suites to eliminate inconsistencies and anomalies.

Disclaimer: While we strive for absolute accuracy, financial data is subject to provider errors; users should cross-reference critical values with official regulatory filings.

Fundamental Data

Our fundamental financial metrics are calculated on a rolling basis utilizing the most recent 12 to 16 quarters of historical data. Financial profiles are updated daily to reflect the latest quarterly earnings releases and regulatory filings.

Market Data

Real-time market data, pricing, and volume metrics are dynamically refreshed every hour to capture intraday market movements.

Quantitative Models & Logic

Rigorous statistical and financial formulas underpinning the screener.

A.

Modified Z-Score

Outlier Resistant

$$M_i = \frac{0.6745 \times (X_i - \tilde{X})}{\text{MAD}}$$

Our scoring methodology utilizes the median and Median Absolute Deviation (MAD) to remain highly resilient against outliers. To standardize metrics, the raw deviation is scaled by a normal distribution adjustment factor ($0.6745$) and strictly winsorized at a threshold of $\pm3.5$. Benchmarking cohorts group companies by geographic region and sector to control for inherent regional and industry volatility. This statistical framework successfully standardizes over 100 disparate corporate fundamental metrics into a single, cohesive, and comparable composite score.

B.

Implied Growth Rate

Iterative Solver

$$\text{EV} = \sum_{t=1}^5 \frac{\text{FCFF}_0 \times (1 + g)^t}{(1 + \text{WACC})^t} + \frac{\text{FCFF}_0 \times (1 + g)^5 \times (1 + g_{\text{terminal}})}{(\text{WACC} - g_{\text{terminal}}) \times (1 + \text{WACC})^5}$$

Variables & Input Parameters

$\text{EV}$: Enterprise Value, reflecting total USD-equivalent market valuation of equity and net debt.
$\text{FCFF}_0$: Current period Free Cash Flow to the Firm (Operating Cash Flow TTM + After-tax Interest Expense - Capex TTM).
$\text{WACC}$: Weighted Average Cost of Capital, reflecting the cost of equity (via CAPM) and after-tax cost of debt.
$g$: Implied Growth Rate, solved iteratively to equate the discounted cash flows to the active enterprise value.
$g_{\text{terminal}}$: Terminal Growth Rate (fixed at $2.5\%$), representing the constant growth rate after year 5.

This calculation reverse-engineers the standard two-stage Discounted Cash Flow (DCF) model to extract the growth rate implied by the current market pricing. By solving for the growth rate that equates the present value of future cash flows to the enterprise value, investors can evaluate whether the market's growth assumptions are realistic. The underlying calculations are processed in Python using scipy and numpy libraries. The calculations are executed systematically by the command-line entry script run_marketanalytics.py.

C.

Relative Discount/Premium

OLS Cohort Model

$$\ln\left(\frac{\text{EV}}{\text{IC}}\right)_i = \beta_0 + \beta_1 \cdot \text{ROIC}_i + \beta_2 \cdot \text{CAGR}_{3\text{yr}, i} + \beta_3 \cdot \text{OCF/Rev}_i + \beta_4 \cdot \text{AssetTurnover}_i + \beta_5 \cdot \text{CII}_i + \beta_6 \cdot \text{WACC}_i + \beta_7 \cdot \ln(\text{MCap}_i) + \epsilon_i$$

Regression Variables & Coefficients

$\ln(\text{EV}/\text{IC})_i$: Natural log of Enterprise Value to Invested Capital for company $i$ (Dependent Variable).
$\beta_0$: Intercept representing cohort base valuation multiple.
$\text{ROIC}_i$: Return on Invested Capital (TTM), measuring profitability relative to investments.
$\text{CAGR}_{3\text{yr}, i}$: 3-Year Compound Annual Revenue Growth Rate, indicating historical growth trend.
$\text{OCF/Rev}_i$: Operating Cash Flow to Revenue ratio, measuring cash generation efficiency.
$\text{AssetTurnover}_i$: Asset Turnover ratio (TTM), reflecting operational asset efficiency.
$\text{CII}_i$: Capital Intensity Index (TTM Capex / Invested Capital).
$\text{WACC}_i$: Weighted Average Cost of Capital, acting as the hurdle rate control.
$\ln(\text{MCap}_i)$: Natural log of USD Market Cap, controlling for size effects.
$\epsilon_i$: Residual error term representing company-specific relative discount or premium.

This regression-based framework determines relative valuation mispricing by comparing a company's actual valuation multiple to its peer-predicted value. The model estimates the OLS regression in log-space across regional and sector cohorts, using residuals to isolate stock-specific discounts or premiums relative to fundamental peers. The core calculations utilize standard statsmodels and scikit-learn Python libraries for cross-sectional regression analysis.

Giraffe Research Documentation