Reproduce “The Only Edge Was Memorized in Training: LLM Trading Agents Around Earnings and Bad News” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Jackson Wang

The authors test whether agents can trade around earnings announcements and bad-news stock drops better than simple executable bets. Across 598 announcements and 409 drops, five models trade on information available at decision time, then face simple controls at fill prices. Agents do not reliably improve; pre-cutoff earnings gains suggest memory, but this does not explain every post-news name effect.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.