FactSet recently hosted a series of expert panels to discuss the trends, challenges, and opportunities shaping the strategies and workflows of the APAC financial community. In a session focused on the ever-changing trading desk, the panel discussed how traders can adapt quickly and prepare for future trends and requirements, including how they can leverage emerging technologies and the move towards automation to achieve best execution. The panel featured Winter WM Chan, Head of Trading Technology APAC at JP Morgan Asset Management, Ferris Kwan, Senior VP Technology at GIC Private Limited, Manu Sharma, Trading Solutions Head of APAC Sales at FactSet, and Kendall Tse, Trader at Vontobel Asset Management.
![apac-focus-panel](https://www.factset.com/hubfs/1)Insight/2021/07.2021/07.29.2021_Need_for_Speed/apac-focus-panel.jpg)
Changes to the trading desk are not a new phenomenon, but with an increasing number of shocks at a rapid pace, traders around the world find themselves with a “need for speed.” A critical part of the investment lifecycle and quintessential in alpha generation, traders have been perceived as individuals grounded in front of giant monitors working in a high-octane office environment. As the past few years have redefined the new normal across traveling, parenting, and working, trading is no exception. Working from home is now a permanent option for traders as they are provided with high spec laptops from their organizations.
Technology teams have worked tirelessly behind the scenes to account for virtual private network (VPN) issues and load balancing to ensure that the environment at home is comparable to the one at work. While there are factors outside of their control, such as the vagaries of home Wi-Fi, adequate measures have been taken to solve for data privacy, cybersecurity, etc., to ensure that traders have access to the right infrastructure to perform with the same precision as they would in the office.
Fragmentation of Liquidity
The trading ecosystem has been going through fundamental changes. A couple decades past, trading was straightforward, with a centralized exchange and the bulk of trading going through that exchange. In 2021, there is a fragmentation of liquidity with various venues in which to trade along with multiple sources of liquidity. In the quest for the best execution, traders need to tap into liquidity from all available sources.
The fragmentation of liquidity persists as there continue to be even more sources; the use of these pools is on the rise and it’s important to understand which ones add value to the firm’s strategy and the end clients. Accessing multiple liquidity sources isn’t enough—buy-side traders need to understand and be comfortable with the type of flows they interact with within these pools as other players such as brokers, prop desk, and high-frequency traders have also entered some of these pools in recent years.
Analyzing Liquidity Is Key
All of this has underlined the importance of data and analytics for traders. They don’t just need access to liquidity, they need quality analysis of the liquidity from a variety of different lenses. Being in close contact with brokers is paramount for a trader as this is where they gauge the pulse of the market. Thus, in the context of analyzing liquidity, it is essential to try to incorporate liquidity coming from brokers.
Indications of interest (IOIs) are a common way for brokers to provide market color to the buy side and it is now common for traders to consume this from all their brokers. Buy-side firms analyze IOIs across the different attributes that need to be accounted for, such as timing and hit rate, and adequate weighting must be assigned for this to provide the right decision support. Models need to continuously learn—how the trader is reacting to the model is also considered an input for there to be improvement. There continue to be many sources of liquidity and this means the traders need to get data quickly and provide actionable intelligence based on that data.
Managing Large Volumes of Data
It is evident that we are living in the era of data and traders are flooded with data—there could potentially be terabytes of data across orders, market data, IOI, pre-trade data, signals, etc. The key to staying ahead is organizing this data so that it provides the right information. There needs to be close collaboration between research (data scientists) and the traders so that the data can be conditioned, filtered, and aggregated before it reaches the traders so that they can make decisions quickly. It is an ongoing balancing act between too much or too little data.
Data visualization is increasingly considered essential. Traders partners with their execution management system (EMS) providers to continually find better ways to visualize data.
At the end of the day, traders are creatures of habit—some traders prefer having access to lots of data as long as it’s clean, structured, and prioritized, as well as easily accessible. On the other hand, other firms prefer to consolidate information, signals, and data in one place so they can apply artificial intelligence (AI) and machine learning (ML) to provide the trader with a recommendation and a summary allowing them to react to the important signals quickly.
Applications of AI and ML
Naturally, when there is too much data, and one hits the limit of analyzing said data, one looks to AI and ML to better scrutinize data and predict results or trends about data that a human might overlook. There are various applications of AI and ML on a trading desk depending on the problem you are looking to solve. One such application can be a broker algo (algorithmic) recommendation which takes in various inputs such as market conditions, historical trade performance, and more to predict the broker or the algo for a specific order under specific conditions. The model analyzes the factors and makes a recommendation when the order arrives. Traders are provided recommendations and are required to provide feedback. The action of the trader and subsequent result of the trade also serve as inputs to the algorithm for subsequent trades, which helps the model continuously improve. The goal here isn’t to replace a trader with AI but merely to augment the trader with better decision support using AI.
Other applications include the analysis of various sources of pre-trade data, such as IOIs or pre-trade signals, which helps filter out the noise by providing fewer recommendations with a higher probability of success. It is also essential to point out that as the adoption of AI increases, so does the regulatory oversight. Firms need to be cognizant that if there is a regulatory review, they are still accountable to explain why a particular strategy and/or broker was used in that situation.
Conclusion
If there is one constant, it is change—and one can expect continued disruptions on the trading desk. It is therefore paramount for organizations to work even more closely with their EMS providers to continuously evolve. Buy-side firms continue to look for better ways to integrate data and liquidity sources in a simple EMS across asset classes and have access to trade analytics which assist traders with navigating market microstructure and liquidity, have a better understanding of fragmentation.
You can listen to the full conversation here. Click here to learn more about FactSet’s front office solutions.
The information contained in this article is not investment advice. FactSet does not endorse or recommend any investments and assumes no liability for any consequence relating directly or indirectly to any action or inaction taken based on the information contained in this article.