Democratizing Quantitative Trading in African Markets: A FinGPT-Based Approach

Victor Jotham Ashioya

Keywords: Quantitative trading, Large Language Models, Emerging Markets, Sentiment Analysis, Financial Inclusion

TL;DR: We built a low-cost trading platform using a resource-efficient Large Language Model to make sophisticated quantitative finance accessible in African markets.

Abstract:

This paper presents FinGPT Trader, a novel confidence-
weighted sentiment analysis system designed to democra-
tize quantitative trading in African markets. The project ad-
dresses significant barriers to entry, including expensive in-
frastructure, limited technical expertise, and restricted ac-
cess to sophisticated trading tools. By leveraging a fine-
tuned Falcon-7B Large Language Model for financial sen-
timent analysis and integrating it with lightweight technical
analysis, FinGPT Trader offers a resourceefficient solution
tailored for environments with constrained resources. Pre-
liminary results indicate significant improvements in acces-
sibility and cost-effectiveness compared to traditional trad-
ing platforms. This approach has the potential to unlock
quantitative trading opportunities for Small and Medium-
sized Enterprises (SMEs), retail investors, and emerging
fund managers across Africa, fostering greater financial in-
clusion and economic development in the region

Submission Number: 1