15 Pages Posted: 27 Apr 2015 Last revised: 5 Jul 2015
Date Written: April 25, 2015
The proliferation of false discoveries is a pressing issue in Financial research. For a large enough number of trials on a given dataset, it is guaranteed that a model specification will be found to deliver sufficiently low p-values, even if the dataset is random.
Most academic papers and investment proposals do not report the number trials involved in a discovery. The implication is that most published empirical discoveries in Finance are likely to be false. This has severe implications, specially with regards to the peer-review process and the Backtesting of investment proposals.
We make several proposals on how to address these problems.
Keywords: Multiple testing, selection bias, backtest overfitting, p-values
JEL Classification: G0, G1, G2, G15, G24, E44
Suggested Citation: Suggested Citation
Lopez de Prado, Marcos, Illegitimate Science: Why Most Empirical Discoveries in Finance Are Likely Wrong, and What Can Be Done About It (Presentation Slides) (April 25, 2015). Available at SSRN: https://ssrn.com/abstract=2599105 or http://dx.doi.org/10.2139/ssrn.2599105