SNB researchers test LLM-based FX trading strategy
Meta’s Llama 3.1 comes out on top at predicting G10 currency sentiment based on news articles.
Researchers from the Swiss National Bank have shown how a trading strategy that uses fine-tuned large language models (LLMs) to analyze sentiment in the foreign-exchange market could outperform traditional language-based artificial intelligence methods.
The authors of the SNB working paper, Daniele Ballinari and Jessica Maly, suggest that by fine-tuning LLMs to better understand the jargon of FX markets, trading strategies can be created based on the sentiment they detect.
LLMs “provide innovative
Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.
To access these options, along with all other subscription benefits, please contact info@waterstechnology.com or view our subscription options here: https://subscriptions.waterstechnology.com/subscribe
You are currently unable to print this content. Please contact info@waterstechnology.com to find out more.
You are currently unable to copy this content. Please contact info@waterstechnology.com to find out more.
Copyright Infopro Digital Limited. All rights reserved.
As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (point 2.4), printing is limited to a single copy.
If you would like to purchase additional rights please email info@waterstechnology.com
Copyright Infopro Digital Limited. All rights reserved.
You may share this content using our article tools. As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (clause 2.4), an Authorised User may only make one copy of the materials for their own personal use. You must also comply with the restrictions in clause 2.5.
If you would like to purchase additional rights please email info@waterstechnology.com
More on Emerging Technologies
Tokenized assets draw interest, but regulation lags behind
Regulators around the globe are showing increased interest in tokenization, but concretely identifying and implementing guardrails and ground rules for tokenized products has remained slow.
CME, LSEG align on market data licensing in GenAI era
The two major exchanges say they are licensing the use case—not the technology.
Waters Wavelength Ep. 341: Citi’s Pitts and Topa
This week, Citi’s Michele Pitts and Marcello Topa join Wei-Shen to talk about UK and EU T+1.
Robeco tests credit tool built in Bloomberg’s Python platform
This follows the asset manager’s participation in Bloomberg’s Code Crunch hackathon in Singapore, alongside other firms including LGT Investment Bank and university students.
Bloomberg expands GenAI summary options on Terminal
The additions include an expansion of its AI-powered news summaries, as well as a new AI summary tool for company-related news content.
AI enthusiasts are running before they can walk
The IMD Wrap: As firms race to implement generative and agentic AI, having solid data foundations is crucial, but Wei-Shen wonders how many have put those foundations in.
Buy-side data heads push being on ‘right side’ of GenAI
Data heads at Man Group and Systematica Investments explain how GenAI has transformed the quant research process.
Jump Trading spinoff Pyth enters institutional market data
The data oracle has introduced Pyth Pro as it seeks to compete with the traditional players in market data more directly.
