A new study published on arXiv explores the effectiveness of admission gates in quantitative strategy research, which are statistical criteria designed to prevent researchers from adopting conclusions based on chance findings. The research introduces an "injected-truth" protocol to test these gates, finding that while they successfully eliminate false discoveries in scenarios with weak signals, they drastically reduce the adoption rate of findings to as low as 1-7%. The study also highlights that using absolute returns instead of excess returns for these criteria can lead to the rejection of all candidate signals, including genuine ones. AI
IMPACT This study highlights potential pitfalls in research methodology that could impact the reliability of AI-driven quantitative strategy findings.
RANK_REASON The item is a research paper published on arXiv discussing a new methodology and findings. [lever_c_demoted from research: ic=1 ai=0.7]
- alphaXiv
- artificial intelligence
- arXiv
- arXivLabs
- Bibliographic Explorer
- CatalyzeX Code Finder for Papers
- Connected Papers
- CORE Recommender
- DagsHub
- Falsification-First Selection
- Gotit.pub
- Hugging Face
- Influence Flower
- Injected-Truth
- Litmaps
- Quantitative Strategy Research
- ScienceCast
- scite Smart Citations
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →