Detection of False Investment Strategies Using Unsupervised Learning Methods

ArXiv ID: ssrn-3167017 “View on arXiv”

Authors: Unknown

Abstract

Most investment strategies uncovered by practitioners and academics are false. This partially explains the high rate of failure, especially among quantitative h

Keywords: quantitative finance, investment strategies, backtesting bias, market efficiency, quantitative strategies

Complexity vs Empirical Score

• Math Complexity: 7.5/10
• Empirical Rigor: 2.0/10
• Quadrant: Lab Rats
• Why: The paper introduces a complex unsupervised learning algorithm involving probability distributions and multiple testing corrections, but lacks specific implementation details, code, or detailed backtesting results, focusing more on theoretical and statistical methodology.

  flowchart TD
    A["Research Goal:<br>Detect false quantitative investment strategies"] --> B["Methodology:<br>Unsupervised Learning (e.g., Clustering)"]
    B --> C["Data Inputs:<br>Strategy Returns, Factor Loadings, Backtest Metrics"]
    C --> D["Computational Process:<br>Identify Outliers & Anomalies in Strategy Space"]
    D --> E["Key Findings:<br>Strategies are often noise; high failure rate due to backtesting bias"]