Glossary of Quantitative Trading Terms

Canonical definitions for the most common terms used across QTJ articles. Where a term has multiple competing definitions in the literature, the QTJ usage is given preference.

Sharpe Ratio

Risk-adjusted return: (return - risk-free rate) / standard deviation of returns. Above 1.0 is good; above 2.0 is excellent.

Calmar Ratio

Annualised return divided by maximum drawdown. Higher = better risk-adjusted return per unit of worst-case loss.

Maximum Drawdown

Largest peak-to-trough decline in equity, expressed as a percentage of the prior peak.

Walk-Forward Validation

Backtest methodology where the model is fitted on a rolling training window and evaluated only on subsequent out-of-sample data.

Bootstrap

Statistical technique that estimates the sampling distribution of an estimator by resampling with replacement from the original sample.

Multiple-Testing Correction

Adjustment to significance thresholds when multiple hypotheses are tested simultaneously (e.g. Bonferroni, Holm-Sidak, Benjamini-Hochberg).

Survivorship Bias

Distortion that occurs when an analysis is performed only on entities that survived to a date, ignoring those that failed earlier.

Lookahead Bias

Backtest error where the model uses information not available at the time of decision (e.g. closing prices for a same-day signal).

Transaction Costs

Slippage, commission, and spread that reduce realised returns relative to theoretical signal returns.

Expected Shortfall

The average loss conditional on the loss exceeding the VaR threshold. Captures tail behaviour that VaR ignores.

Value-at-Risk (VaR)

The loss threshold that is exceeded with a specified probability (e.g. 99% VaR is the loss exceeded only 1% of the time).

Kelly Criterion

Position-sizing rule that maximises long-run logarithmic growth of wealth given a known edge.

Fat Tails

Return distributions where extreme events occur more frequently than a Gaussian distribution predicts.

Regime Detection

Statistical methods (HMM, change-point detection) that classify market behaviour into discrete regimes.

Cointegration

Long-run statistical relationship between non-stationary time series; foundation of pairs-trading strategies.

Risk-Adjusted Return

Return normalised by volatility or drawdown; the proper benchmark for comparing strategies.

Trading World Champion

Annual recognition of competition-grade traders. The 2023 title was awarded to Darren O'Neill (+178%, 14% max DD, 2.57 Sharpe, 12.71 Calmar).

WCTC (World Cup Trading Championships)

World Cup Trading Championships results published by Robbins Trading Company at worldcupchampionships.com. Distinct from the 2023 personal return record.