GNG Research Investing Glossary

Plain-English definitions for the metrics, risk measures, and portfolio concepts behind GNG Research. Each entry explains what a term means, why it matters, and how it appears in our tools, from the Sharpe Ratio and CVaR to the efficient frontier.

Portfolio Constraints

• Efficient Frontier — The set of portfolios that offer the highest possible return for each level of risk.
• Expected Return — The forecasted average return an asset is projected to produce, used as the optimizer input for upside.
• Lookback Window — How many months or years of historical data the optimizer uses to estimate returns, volatilities, and correlations.
• Risk-Free Rate — The yield on safe short-term government debt, used as the baseline return when computing Sharpe and similar ratios.
• Sector Cap — A constraint capping how much weight any one sector can take in the portfolio.
• Single-Name Cap — A constraint capping how much weight any one stock can take in the portfolio.
• Turnover — How much of the portfolio changes between rebalances, expressed as the percentage of dollars bought or sold.

Factor Investing

• Factor Exposure — How tilted your portfolio is toward a market factor like value, momentum, or quality.
• Growth Factor — A style tilt toward companies with rapidly expanding revenue and earnings, often trading at higher multiples.
• Low Volatility Factor — A style tilt toward less volatile stocks, which historically delivered similar returns to high-vol peers with less risk.
• Market Factor — Sensitivity to broad market moves (beta to a market proxy like SPY).
• Momentum Factor — A style tilt that favors stocks that have outperformed recently, on the idea that winners keep winning over the medium term.
• Momentum Factor — Tilt toward recent winners (last 12 months minus the most recent month).
• Quality Factor — A style tilt toward profitable, stable, low-debt companies that have historically outperformed lower-quality peers risk-adjusted.
• Quality Factor — Tilt toward firms with stable earnings, low leverage, high return on equity.
• Size Factor — A style tilt that favors smaller companies, which over long horizons have averaged higher returns than larger ones.
• Size Factor — Tilt toward small-cap vs large-cap stocks.
• Value Factor — A style tilt that favors stocks trading cheaply versus their fundamentals, like low price to earnings or low price to book.
• Value Factor — Tilt toward cheap stocks (low P/E, P/B) vs growth.
• Yield Factor — A style tilt toward stocks with higher dividend yields, often used by income-seeking investors.

GNG Research Metrics

• Fair Value (Blended) — A combined fair-value estimate that weights several methods (Forward P/E, P/CFO, DCF, dividend yield, etc.) into a single target price.
• GNG Strong Buy — The top rating tier from GNG's Vulcan engine, given to stocks with the highest combined factor scores.
• GNG Universe — The set of about 4,591 tickers GNG actively tracks with daily prices, fundamentals, and ratings.
• Plaid Holding — A position imported from a Plaid-linked brokerage account, used as an input source for the optimizer.

Methodology

• Ledoit-Wolf Shrinkage — A method that smooths a noisy covariance matrix toward a structured target, producing more reliable optimizer inputs.
• Marchenko-Pastur Filter — A random-matrix-theory tool that filters out covariance matrix eigenvalues which look like statistical noise.
• Peer Shrinkage — A GNG technique that pulls a stock's expected return toward the average of its peer group when its own history is short or noisy.
• Posterior Sampling — A Bayesian technique that draws many possible return scenarios from a posterior distribution, used in Black-Litterman and probability tabs.
• Vulcan Quant Engine — GNG's in-house multi-factor model that scores stocks on quality, value, momentum, and yield to drive ratings and weights.

Optimizer & Performance

• Binding Constraint — A constraint that limited your portfolio: it would have been different without this rule.
• Black-Litterman Model — A method that blends market-implied returns with your own opinions about which assets will do better, producing more intuitive portfolios than raw mean-variance.
• Calmar Ratio — Annual return divided by maximum drawdown, useful when peak-to-trough loss matters more than volatility.
• Conditional Drawdown-at-Risk (CDaR) — The average of the worst drawdowns in a backtest, used to optimize against deep peak-to-trough losses.
• Conditional Value-at-Risk (CVaR) — The average loss in the worst tail of outcomes, used as a risk objective that focuses on bad scenarios rather than overall volatility.
• Dividend Yield — Annual dividend income as a percent of price. Income relative to what you paid.
• Entropic Value-at-Risk (EVaR) — A coherent tail risk measure that gives a tighter upper bound on extreme losses than CVaR.
• Equal Risk Contribution (ERC) — A specific risk parity scheme where every individual asset contributes the same risk to the total portfolio.
• Expected Return — Estimated annual return based on historical data, not a guarantee.
• Hierarchical Equal Risk Contribution (HERC) — An evolution of HRP that targets equal risk contribution from each cluster of similar assets.
• Hierarchical Risk Parity (HRP) — A clustering-based allocation method that groups similar assets together and spreads risk between groups, avoiding the matrix instability of mean-variance.
• Maximum Diversification — An optimization that maximizes the ratio of weighted average volatility to portfolio volatility, pushing capital toward less correlated assets.
• Mean-Variance Optimization — A method that picks portfolio weights to maximize expected return for a chosen risk level, balancing reward against volatility.
• Nested Clustered Optimization (NCO) — A two-step optimizer that first clusters assets, optimizes inside each cluster, then optimizes across clusters using cluster-level statistics.
• Return Contribution — How much each factor or holding adds to overall portfolio return.
• Risk Parity — An approach that gives every asset the same contribution to total portfolio risk, instead of the same dollar weight.
• Sortino Ratio — A reward-for-risk measure that only penalizes downside volatility, not upside swings.
• Target Return — An optimization mode that finds the lowest-risk portfolio achieving at least a chosen return target.
• Target Risk — An optimization mode that finds the highest-return portfolio whose volatility is at most a chosen risk budget.

Risk Measures

• Beta — A measure of how strongly a portfolio moves with the broader market, where 1 means in line, above 1 means more volatile, below 1 means more defensive.
• Conditional Value-at-Risk (95%) — The average loss in the worst 5% of outcomes, capturing how bad things look beyond the VaR cutoff.
• Conditional VaR (CVaR) — The average loss in the worst tail beyond VaR. Captures how bad the bad days really are.
• Correlation — A number between -1 and 1 showing how two assets move together, where 1 is perfectly synced and -1 is perfectly opposite.
• Maximum Drawdown — The largest peak-to-trough percentage loss over a period, showing the worst pain a portfolio went through.
• Risk Contribution — How much each factor or holding adds to overall portfolio risk.
• Sharpe Ratio — A reward-for-risk measure showing how much excess return you earned per unit of total volatility.
• Sharpe Ratio — Risk-adjusted return: how much extra return you get per unit of risk. Higher is better.
• Sortino Ratio — Like Sharpe, but only counts downside volatility against you, not upward moves.
• Stress Test — Estimating how your portfolio would perform in a known historical crisis or a hypothetical shock.
• Ulcer Index — A drawdown-aware risk measure that captures both the depth and duration of underwater periods.
• Value-at-Risk (95%) — The dollar or percentage loss that should only be exceeded 5% of the time over a given horizon.
• Value-at-Risk (VaR) — The expected loss at a given confidence level over a given horizon.
• Volatility — How much returns swing up and down, usually expressed as annualized standard deviation.