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

TL;DR

This benchmark study uses state-of-the-art time series models including pre-trained LLMs (GPT-2 as backbone), transformers, and linear models to demonstrate their ability to outperform traditional approaches, even with minimal ("few-shot") or no fine-tuning ("zero-shot").

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.

Figures (2)

• Figure 1: Financial Aid data aggregated at the state level from 2004 to 2020 (17 years), in billions of US dollars. Access to historical datasets is limited to yearly intervals.
• Figure 2: A high-level overview of pre-training an LLM and fine-tuning on a custom dataset (e.g. the Financial Aid dataset) for downstream tasks.