Large Language Models for Financial Aid in Financial Time-series Forecasting

TL;DR

This paper explores using large language models, including GPT-2, to improve financial time series forecasting in financial aid, especially with limited data, demonstrating their effectiveness over traditional methods.

Contribution

It introduces a benchmark study showing that pre-trained LLMs can outperform traditional models in financial aid time series forecasting with scarce data.

Findings

LLMs outperform traditional models in scarce data scenarios

Pre-trained models achieve high accuracy with minimal fine-tuning

Financial aid time series can benefit from advanced NLP models

Abstract

Considering the difficulty of financial time series forecasting in financial aid, much of the current research focuses on leveraging big data analytics in financial services. One modern approach is to utilize "predictive analysis", analogous to forecasting financial trends. However, many of these time series data in Financial Aid (FA) pose unique challenges due to limited historical datasets and high dimensional financial information, which hinder the development of effective predictive models that balance accuracy with efficient runtime and memory usage. Pre-trained foundation models are employed to address these challenging tasks. We use state-of-the-art time series models including pre-trained LLMs (GPT-2 as the backbone), transformers, and linear models to demonstrate their ability to outperform traditional approaches, even with minimal ("few-shot") or no fine-tuning ("zero-shot"). Our benchmark study, which includes financial aid with seven other time series tasks, shows the potential of using LLMs for scarce financial datasets.