Abstract
The rate of failure in quantitative finance is high, particularly in financial machine learning applications. The few managers who succeed amass a large amount of assets and deliver consistently exceptional performance to their investors. However, that is a rare outcome, for reasons that the author explains in this article. In the author's experience, 10 critical mistakes underlie those failures.
- © 2018 Pageant Media Ltd
Don’t have access? Register today to begin unrestricted access to our database of research.
Log in using your username and password
In this issue
The 10 Reasons Most Machine Learning Funds Fail
Jump to section
- Article
- Abstract
- PITFALL #1: THE SISYPHUS PARADIGM
- SOLUTION #1: THE META-STRATEGY PARADIGM
- PITFALL #2: RESEARCH THROUGH BACKTESTING
- SOLUTION #2: FEATURE IMPORTANCE ANALYSIS
- PITFALL #3: CHRONOLOGICAL SAMPLING
- SOLUTION #3: THE VOLUME CLOCK
- PITFALL #4: INTEGER DIFFERENTIATION
- SOLUTION #4: FRACTIONAL DIFFERENTIATION
- PITFALL #5: FIXED-TIME HORIZON LABELING
- SOLUTION #5: THE TRIPLE-BARRIER METHOD
- PITFALL #6: LEARNING SIDE AND SIZE SIMULTANEOUSLY
- SOLUTION #6: META-LABELING
- PITFALL #7: WEIGHTING OF NON-INDEPENDENT IDENTICALLY DISTRIBUTED SAMPLES
- SOLUTION #7: UNIQUENESS WEIGHTING AND SEQUENTIAL BOOTSTRAPPING
- PITFALL #8: CROSS-VALIDATION LEAKAGE
- SOLUTION #8: PURGING AND EMBARGOING
- PITFALL #9: WF (OR HISTORICAL) BACKTESTING
- SOLUTION #9: COMBINATORIAL PURGED CROSS-VALIDATION
- PITFALL #10: BACKTEST OVERFITTING
- SOLUTION #10: THE DEFLATED SHARPE RATIO
- CONCLUSIONS
- ACKNOWLEDGMENT
- REFERENCES
- Info & Metrics
Similar Articles
Cited By...
- Interactions and Interconnectedness Shape Financial Market Research
- A Network and Machine Learning Approach to Factor, Asset, and Blended Allocation
- Machine Learning in Asset Management--Part 1: Portfolio Construction--Trading Strategies
- AI and Machine Learning in Real Estate Investment
- A Backtesting Protocol in the Era of Machine Learning