Houde Qian, Zheng Jiang, WenYi Li, Chenxi Xu, Xinjie Lin, Yuzhao Peng, Ruoyu Han, Youjie Zheng, Kaisen Yang, Qinhuai Na, You Fu, Calvin Xiao · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.30113 · cited by 0 works (OpenAlex) · open-access PDF
A pesquisa quantitativa é um ciclo iterativo: um pesquisador propõe uma hipótese de mercado, implementa-a como um sinal executável, testa-a em dados históricos e decide o que manter. Os benchmarks existentes pontuam fórmulas, código ou decisões comerciais individuais, deixando aberto se os grandes modelos de linguagem (LLMs) podem completar esse ciclo e produzir estratégias que funcionam além do período de pesquisa. Introduzimos o Quant-Arena, um benchmark e um painel de liderança público para pesquisas quantitativas de fim a fim sobre dados do mercado real A-share. Um modelo constrói características e código de pontuação de ações, recebe feedback apenas em 2022-2023 e congela uma estratégia para avaliação no ano selado 2024. Um avaliador de conta de caixa aplica restrições
Machine translation (Meta M2M100); the English original follows.
Quantitative research is an iterative loop: a researcher proposes a market hypothesis, implements it as an executable signal, tests it on historical data, and decides what to keep. Existing benchmarks score formulas, code, or individual trading decisions, leaving open whether large language models (LLMs) can complete this loop and produce strategies that work beyond the research period. We introduce Quant-Arena, a benchmark and public leaderboard for end-to-end quantitative research on real A-share market data. A model constructs features and stock-scoring code, receives feedback only on 2022--2023, and freezes one strategy for assessment on the sealed year 2024. A cash-account evaluator applies trading constraints and costs, including market impact estimated from a simulated order book. Scores are dominated by realized excess return and complemented by signal quality and the diversity of features used by the strategy. We evaluate nine frontier models, among which GPT-6 Astra scores highest; yet no model beats the equal-weight market in 2024. Quant-Arena is open to submissions from any model or agent harness.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Benchmarking, Executable, Benchmark (surveying), Computer science, Order (exchange), Algorithmic trading
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.30113, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange