| Author/Name | Definition |
| Panel A: Lexical definitions | |
| Burgess (2000) | SA is a framework for identifying, modelling and exploiting small but consistent regularities in asset price dynamics |
| Zapart (2003) | SA is an investment opportunity arising from the choice of models for hedging |
| Do et al. (2006) | SA is an equity trading strategy that employs time series methods to identify relative mispricing between stocks |
| Thomaidis and Kondakis (2006) | SA is an attempt to profit from pricing discrepancies that appear in a group of assets |
| Pole (2007) | SA uses mathematical models to generate returns from systematic movements in securities prices |
| Avellaneda and Lee (2008) | SA encompasses a variety of strategies characterized by: i) systematic trading signals, ii) market neutral trades and iii) statistical methods |
| Montana et al. (2008) | SA is an investment strategy that exploits patterns detected in financial data streams |
| Panel B: Conceptual definitions | |
| Connor and Lasarte (2003) | SA is a zero-cost portfolio where the probability of a negative payoff is very small but not exactly zero |
| Stefanini (2006) | SA seeks to capture imbalances in expected value of financial instruments, while trying to be market neutral |
| Saks and Maringer (2008) | SA accepts negative pay-outs with a small probability as long as the expected positive payouts are high enough and the probability of losses is small enough |
| Focardi, Fabozzi and Mitov (2016) | SA strategies aim at producing positive, low-volatility returns that are uncorrelated with market returns |
| Panel C: Operational definitions | |
| δ-Arbitrage | Is a strategy with a Sharpe ratio above a constant and positive δ |
| Good Deal | Consists in buying (selling) securities whose market price lies outside a range of plausible prices |
| Approximate Arbitrage | Is a strategy whose gain-loss ratio is above a predefined constant value greater than 1 |
| Acceptable Opportunity | Is a strategy with a non-negative expected value under each valuation measure and losses capped under the set of stress measures |
| ε-Arbitrage | Consists in buying (selling) those derivatives strategies whose price significantly differs from the least costly optimal replication strategy |
| SA (Bondarenko) | Is a strategy with expected positive payoff and expected non-negative payoff conditional on the augmented information set |
| SA (Hogan et al.) | With time the strategy has positive expected payoff, probability of a loss which tends to zero and time averaged variance which converges to zero |