Detection of False Investment Strategies Using Unsupervised Learning Methods
28 Pages Posted: 23 Apr 2018 Last revised: 25 Jun 2018
Date Written: April 22, 2018
Most investment strategies uncovered by practitioners and academics are false. This partially explains the high rate of failure, especially among quantitative hedge funds (smart beta, factor investing, stat-arb, CTAs, etc.) In this paper we examine why false positives are so prevalent in finance, why researchers fail (in many cases purposely) to detect them, and why firms are able to monetize their scheme. Beyond merely pointing to this industrywide problem, we offer a practical solution. We hope that the machine learning tools presented in this paper will help financial academic journals filter out false positives, and bring up the retraction rate to reasonable levels. The SEC, FINRA and other regulatory agencies worldwide could use these tools to take a more active role in curving this rampant financial fraud.
A presentation based on this paper can be found at https://ssrn.com/abstract=3173146
Keywords: Backtest overfitting, selection bias, multiple testing, quantitative investments, machine learning, financial fraud
JEL Classification: G0, G1, G2, G15, G24, E44
Suggested Citation: Suggested Citation