Paper: arXiv 2603.09303

Authors: Hanyong Cho, Geumil Bae, Jang Ho Kim

Abstract

This paper investigates how large language models (LLMs) form and express investor risk profiles, a critical component of retail investment advising. We examine three LLMs (GPT, Gemini, and Llama) and assess their responses to a standardized risk questionnaire under varying prompts. In particular, we establish each model’s default investment profile by analyzing repeated responses per model. We observe that LLMs are generally longterm investors but exhibit different tendencies in risk tolerance: Gemini has a moderate risk level with highly consistent responses, Llama skews more conservative, and GPT appears moderately aggressive with the greatest variation in answers. Moreover, we find that assigning specific personas such as age, wealth, and investment experience leads each LLM to adjust its risk profile, although the extent of these adjustments differs across the models.

Complexity vs Empirical Score

  • Math Complexity: 2.0/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper provides an empirical investigation into how main-stream LLMs evaluate investor risk profiles using established statistical tests (ANOVA and Kruskal-Wallis). While lacking advanced mathematical derivations, it executes a clean, reproducible framework evaluating model default behaviors and response variations to persona prompting.

Research Flowchart

  flowchart TD
    A[Research Goal: Assess Investor Risk Profiles & Persona Sensitivity in LLMs] --> B[Input: Standardized Financial Risk Questionnaire]
    
    B --> C[Experiment 1: Prompt Default Models <br>GPT, Gemini, Llama repeatedly]
    B --> D[Experiment 2: Persona Prompting <br>Vary Age, Wealth, & Experience]
    
    C & D --> E[Quantitative Response Analysis]
    
    E --> F[Default Profiles: Gemini=Moderate/Consistent, Llama=Conservative, GPT=Mod-Aggressive/Varied]
    E --> G[Persona Impact: All LLMs adapt risk profiles to personas, but with varying sensitivity]