THETRADETECHFX DAILY
When it comes to FX what trends are at the forefront when trading on an alpha mandate? We trade FX primarily in the context of our Global Data Sentiment strategy. In this strategy, we systematically process more than 1.5 million newspaper articles every day and derive trading signals from their underlying sentiment. We do this for stock index and bond futures, but also for foreign currencies. Currently, this is only done for developed market currencies, as the number and quality of available news for emerging market currencies is rather low. However, our analysis shows that this is changing for the better, so we expect to expand our investment universe to include emerging market currencies in the coming years. In addition, we have observed that sentiment trends in currencies have become shorter in recent times. In our view, this is due to the fact that markets are currently being driven primarily by political decisions. If this trend continues, we will have no choice but to adjust both our signals and our trading frequency.
Which innovations would you highlight as key for the buy-side going forward? We are currently focusing our research on the application of large language models( LLMs) to determine the sentiment of newspaper articles. The current state of the art is the use of psychological dictionaries to determine sentiment. This involves counting the positive and negative connotated words in an article and comparing them to each other. However, this approach has the disadvantage that it does not take the context of an article into account, e. g., whether the outlook is more positive or negative. Our analysis shows that LLMs can help here, as they are also able to capture the respective context. We expect to be actively using LLMs in our sentiment analyses by the end of this year.
Many believe technology to be the key to unlocking FX’ s full potential. Are the skillsets of traders themselves therefore changing also? We focus on generating alpha, including foreign currency trading. A lot has happened technologically in this area in recent years. For example, parallel data processing in the cloud enables us to process millions of newspaper articles every day. We believe this trend will continue, allowing us to expand our database beyond newspaper articles. The skillset of quantitative asset managers is changing in such a way that efficient data processing and data scouting are becoming increasingly important. So, the
LLMs set to play an increasingly important role in FX traders’ day-to-day
The TRADE sits down with MARKUS EBNER, head of multiasset at Quoniam Asset Management, to unpack the trends making the most impact on FX trading, delving into what’ s front of mind when generating alpha, the impact AI is set to have on established processes, and how all this change is leading to an evolution in traders’ skillsets going forward.
skillset is definitely getting more technical than it was in the past.
Looking to the future, how are FX desks set to look different in 3, 5, 10 years’ time? In the medium term, the trend in quantitative portfolio management will continue, with more and more data being processed in ever shorter periods of time. For us, this means that we will be processing more newspaper articles or alternative data sets such as company publications. The use of LLMs will also play an increasingly important role. These are powerful tools that may one day enable us to read between the lines of articles. In the long term, I believe that artificial intelligence can relieve us of a lot of technical work. It already helps us considerably in the creation of programming code. In the long term, AI will probably be able to take over this task completely.
The official newspaper of TradeTech FX 2025 TheTradeNews. com 17