algorithmic-trading

FinLLM-KG 是一个模型无关的框架，将因果 LLM 与金融知识图谱结合，从 SEC 10-K 文件中注入公司级、时间对齐的知识以改进金融情绪分析与算法交易，同时提供可解释的预测依据。

该论文提出一种将情感分析深度融入Actor-Critic类深度强化学习智能体（A2C、PPO、DDPG、TD3、SAC）的股票交易策略，发现结合Q学习的策略优化方法在融合情感信息后能显著提升盈利与风险规避能力。

This paper studies how to specialize large language models for generating executable algorithmic trading code, showing that framework-oriented continued pretraining plus supervised fine-tuning on validated request-to-code pairs greatly improves success on the QuantCode-Bench benchmark and agentic strategy generation, while also revealing a capability-retention failure where domain specialization degrades structured tool calling.

This study evaluates whether test-time reasoning in large language models (like DeepSeek, GPT, and Gemini) improves net portfolio returns in trading, finding that additional reasoning does not reliably enhance economic outcomes across various conditions.

Indicator Go is a Go library that provides 80+ technical analysis indicators, a backtesting framework, and integrates with AI tools via MCP for algorithmic trading.

AI-Trader is an open-source Python framework that simplifies backtesting and optimizing algorithmic trading strategies across multiple markets using config-driven YAML files and AI assistant integration.

Presents Backtrader-Bench, a benchmark for evaluating LLM coding agents in algorithmic trading using self-generated multiple-choice questions and a generator-solver filtering pipeline, showing tool-augmented agents achieve 90% accuracy.

FinSMART introduces a market-aligned reinforcement learning framework for financial sentiment analysis, optimizing sentiment signals with realized market outcomes and achieving a 220% improvement in cumulative trading returns over the strongest baseline.

This paper studies LLMs for parent-order execution in algorithmic trading, introducing PACE, a hierarchical framework that outperforms traditional baselines on Shenzhen Stock Exchange data and suggests LLMs can complement human traders.

A user curated a free GitHub repository aggregating numerous open-source quantitative finance tools, including pricing engines, backtesting frameworks, order book simulators, and risk models, making institutional-grade research tools accessible to individuals at minimal cost.

This paper investigates whether reinforcement learning can discover price manipulation strategies more effectively than traditional model-based approaches in an Almgren–Chriss framework, finding that RL outperforms model-based methods under parameter uncertainty for intermediate volatility.

Proposes an uncertainty-aware reinforcement learning framework for algorithmic trading that integrates distributional, epistemic, and aleatoric uncertainty using SHAP-weighted reconstruction, MC Dropout, and LSTM consensus. Outperforms traditional models on five major US stock indices.

A University of Johannesburg working paper proposes a Kalman-filtered, momentum-extended Anticor algorithm (K-ACM) that, in a backtest from 1962–1984, produces an extreme return of $1 to $36 billion, though acknowledged as a backtest artifact due to unrealistic assumptions.

Introduces AlgoEvolve, an LLM-driven evolutionary framework that generates and iteratively improves algorithmic trading strategies, with a meta-evolutionary outer loop that evolves prompts to guide the inner loop synthesis.

Explores whether AI-driven trading is feasible and secure, addressing potential risks and benefits.

ai-quant-lab is an open-source Python framework that uses Claude to generate algorithmic trading strategies and rigorously validates them with statistical tests to avoid overfitting.

This article outlines Jim Simons' quantitative trading philosophy and argues that AI tools like Claude Code can lower the barrier for ordinary people to build automated trading systems, enabling strategy validation and compounding.

A 29-year-old Oklahoma sales consultant claims to have built an Ethereum price prediction system using Claude and multiple AI agents, replacing an entire quant team and allegedly generating over $300,000 in monthly profits. The content originates from social media, its authenticity is questionable, and it carries clear signs of marketing promotion.

QuantCode-Bench is a new benchmark of 400 tasks evaluating LLMs on generating executable algorithmic trading strategies for the Backtrader framework, testing syntactic correctness, backtest execution, trade presence, and semantic alignment. The study finds that current models' main limitations lie in correctly operationalizing trading logic and API usage rather than syntax.

This paper proposes an orchestration framework for financial agents that maps traditional algorithmic trading components to AI agents, demonstrating improved returns in stock and BTC trading compared to market benchmarks.