Please rotate your device to landscape mode to view the charts.

Option Return Predictability with Machine Learning and Big Data

Journal: Review of Financial Studies

Date: 20230901

Author: Bali, Turan G.; Beckmeyer, Heiner; Morke, Mathis; Weigert, Florian

Abstract:
Drawing upon more than 12 million observations over the period from 1996 to 2020, we find that allowing for nonlinearities significantly increases the out-of-sample performance of option and stock characteristics in predicting future option returns. The nonlinear machine learning models generate statistically and economically sizable profits in the long-short portfolios of equity options even after accounting for transaction costs. Although option-based characteristics are the most important standalone predictors, stock-based measures offer substantial incremental predictive power when considered alongside option-based characteristics. Finally, we provide compelling evidence that option return predictability is driven by informational frictions and option mispricing. Authors have furnished an Internet Appendix, which is available on the Oxford University Press Web site next to the link to the final published paper online.

Link: Google Scholar

Key Findings

Nonlinear Models Outperform

Nonlinear machine learning models significantly outperform linear models in predicting option returns, with gradient-boosted regression trees achieving out-of-sample R² of 2.26%

Profitable Trading Strategy

Machine learning predictions generate monthly returns of 2.04% even after accounting for transaction costs and margin requirements

Key Predictive Factors

Option contract characteristics and market frictions are the most important predictors, with implied volatility being the single most influential factor

Model Performance Comparison

  • Nonlinear models (GBR, RF, Dart, N-En) consistently achieve positive R² values
  • Linear models all produce negative R² values
  • Nonlinear ensemble (N-En) achieves the highest predictive power at 2.45%

Trading Strategy Returns

  • N-En strategy remains profitable even with high transaction costs
  • Returns decrease as trading costs increase but remain positive
  • N-En consistently outperforms L-En across all cost scenarios

Feature Group Importance

  • Contract-based features are most important, accounting for 35% of predictive power
  • Illiquidity and risk measures are second and third most important
  • Traditional stock characteristics have relatively lower importance

Contribution and Implications

  • First comprehensive study applying machine learning to predict individual option returns using both option and stock characteristics
  • Demonstrates importance of nonlinear interactions in option markets
  • Provides practical framework for implementing profitable option trading strategies
  • Shows that option return predictability is driven by informational frictions and market inefficiencies

Data Sources

  • Model Performance Comparison chart based on Figure 1 and Table 2 from the paper
  • Trading Strategy Returns visualization constructed using data from Table 6
  • Feature Group Importance chart created using data from Figure 7 and related discussion