MétaCan
Menu
Back to cohort

A Multi-Agent Approach to Investor Profiling Using Large Language Models

2025· article· en· W4413018474 on OpenAlexaff
Zijiang Yang

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldDecision Sciences
TopicStock Market Forecasting Methods
Canadian institutionsYork University
Fundersnot available
KeywordsProfiling (computer programming)Computer scienceNatural language processingProgramming language

Abstract

fetched live from OpenAlex

Investor profiling is essential in financial advising, allowing advisors to tailor investment strategies based on individual risk preferences, experience, and financial goals. This research aims to automate and enhance the investor profiling process using large language models (LLMs) through interactive multi-agent conversations. Our approach involves designing an investor agent, which represents a pre-defined investor derived from a narrative generated by a large language model (LLM) with given attributes, and an advisor agent that engages in conversation to infer the hidden attributes of the investor. The advisor-agent dynamically adjusts its questions based on previous conversation context to maximize the accuracy of its attribute predictions. The advisor agent makes predictions once it acquires sufficient information and compares them against the ground truth. We conducted extensive simulations across thousands of investor attribute sets and evaluated the effectiveness of the advisor-agent’s predictions based on key metrics. Our results demonstrate that LLM can effectively approximate investor characteristics. This research contributes to the field of AI-driven financial advising and unveils the potential of conversational agents in refining investor assessment methodologies.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.008
metaresearch head score (Gemma)0.008
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.750
Threshold uncertainty score0.963

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0080.008
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.002
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0010.001
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.309
GPT teacher head0.467
Teacher spread0.158 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same topicStock Market Forecasting MethodsFrench-language works237,207