The Price of Thought: Does Test-Time Reasoning Pay in LLM Trading?
cs.AI
Submitted: 2026-09-25
Updated: 2026-09-25
License: http://creativecommons.org/licenses/by/4.0/
The gist: While inference-time reasoning in large language models (LLMs) promises better decision making, its higher computational cost may not yield better economic outcomes.
Terminology
Abstract
While inference-time reasoning in large language models (LLMs) promises better decision making, its higher computational cost may not yield better economic outcomes. Yet reasoning controls are rarely evaluated as economic interventions, where changes in model outputs must translate into better portfolios after trading costs. We conduct a controlled study of representative LLMs from the DeepSeek, GPT, and Gemini families. We vary reasoning effort while holding information available at each formation date, prompts, output formats, and portfolio construction fixed. Our evaluation covers a full year of U.S. equities under three input conditions: numerical, identifiable news, and masked news. It includes more than 800,000 asset predictions and repeated model generations. Across all three model families, additional reasoning does not produce a reliable improvement in net portfolio returns. For DeepSeek, where we examine the full progression from no reasoning to maximum reasoning, performance is nonmonotonic. Repeated generations also produce unstable treatment effects and portfolio selections, even when overall scores remain similar. These findings show that additional reasoning can change financial decisions without reliably improving their economic value, motivating validation for each task before deployment.
Sources
- FinTradeBench: A Financial Reasoning Benchmark for LLMs
- Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs
- FinBERT: Financial Sentiment Analysis with Pre-trained Language Models
- StockBench: Can LLM Agents Trade Stocks Profitably In Real-world Markets?
- FinQA: A Dataset of Numerical Reasoning over Financial Data
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- INVESTORBENCH: A Benchmark for Financial Decision-Making Tasks with LLM-based Agent
- Profit Mirage: Revisiting Information Leakage in LLM-based Financial Agents
- Let's Verify Step by Step
- Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models
- s1: Simple test-time scaling
- Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters
- When Reasoning Fails: Evaluating 'Thinking' LLMs for Stock Prediction
- Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- BloombergGPT: A Large Language Model for Finance
- Agentic Trading: When LLM Agents Meet Financial Markets
- TradingAgents: Multi-Agents LLM Financial Trading Framework
- FinBen: A Holistic Financial Benchmark for Large Language Models
- PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark for Finance
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems