Investment biases in financial and general language models | |
| Author | Minn Banya |
| Call Number | AIT Thesis no.CS-25-03 |
| Subject(s) | Finance--Mathematical models Finance--Decision making--Data processing Natural language generation (Computer science) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Computer Science |
| Publisher | Asian Institute of Technology |
| Abstract | Language models are increasingly utilized in financial decision-making, wheretheir out puts can directly or indirectly influence investment strategies and outcomes in a major way. As such, the existence of biases within these models—such as Confirmation Bias, Anchoring Bias, Herding Effect, Recency Bias, and Loss Aversion—is a point of concern. This study investigates the existence and extent of these biases in financial language models through the creation of a clean, curated dataset of approximately 1000 samples. By systematically perturbing each sample to represent specific biases, the dataset enables a controlled and focused evaluation of model behavior.Usingthis dataset, two financial language models and two general language modelswere analyzed to uncover patterns of biased predictions and their alignment with known cog nitive biases in investment decision-making. The study also explores potential strategies for bias mitigation, providing insights into reducing bias impact. By offering a novel dataset and detailed analysis, this research contributes to a deeper understanding of investment biases in financial language models and lays the groundwork for improving fairness and transparency in their applications. |
| Year | 2025 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Chaklam Silpasuwanchai |
| Examination Committee(s) | Chantri Polprasert;Attaphongse Taparugssanagorn |
| Scholarship Donor(s) | AIT Scholarship |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2025 |