The Future: Trading Systems Hunt for Alpha
Trading systems of the future will harness predictive technology in order to generate alpha, capitalizing on the data which is generated during the trading cycle but until now had not been utilized.
“We are very focused at Franklin Templeton on what we call ‘alphatizing our technology’ which basically means making our technology work smarter and more predictable in our investment process,” said Bill Stephenson, head of global trading strategy at Franklin Templeton Investments.
“We have been able to leverage our technology team to build a database to collect transactional data, engage our quantitative team to normalize, model, and back-test performance statistics, and educate our traders on how to utilize the results to drive better at-the-trade decision making,” Stephenson said. “This one example has the potential to shave 15% or more off our implicit trading costs when utilizing certain strategies.”
The biggest challenge, he adds, is bridging the knowledge gap between investment, technology, and quantitative personnel. “You really need a team with these diverse skill sets that is incentivized to combine their knowledge base in order to create the intelligence across our systems and processes to generate new alpha,” Stephenson said.
Data capture and utilization is very much a part of this move toward smarter trade execution.
“We have seen a massive trend where data is no longer just important for reporting, but for driving future decisions,” said Alfred Eskandar, CEO of trading systems provider Portware. “Institutions have always had an information advantage which was unused. We are helping portfolio managers identify trends, and put that into the trading process in a systematic fashion that yields consistent results. A trading desk and a portfolio manager are now literally on the same page using the same data to drive consistent performance.”
As part of its commitment to redefine way the buy side trades, Portware provides Alpha Vision, an algorithmic optimization system that uses predictive analytics to help traders ensure that they are in the right algorithm at the right time, all the time. By delivering critical market intelligence and color to traders in real time, Alpha Vision gives traders control over their order flow.
“We are giving the trading desk decision support tools to give them comfort that the decisions they make are playing out as expected, so they can make tweaks if they’re not,” Eskandar said. “Dynamic data is where we live now. Lots of industries are using predictive analytics to fine tune their business modes, but it hasn’t been used in financial services until now. We are automating the trade cycle for the benefit of the institutions, while giving them information and color as its happening. That’s all using historical data, predictive models, and reinforcing them with reports.”
Franklin Templeton Investments hasn’t fully integrated new predictive technology into its process, but is finding ways to leverage its technology and analytics to create more automation.
“So while the trader steers the ship and monitors the course, automation adds a layer of predictability into our process which allows the trader to engage more deeply into the most opportunistic investment scenarios,” Stephenson said.
For example, this may mean taking lower-touch trade flow and combining it with an automated decision processor utilizing our back-tested strategies.
“Some consider portfolio manager or trader alpha profiling a form of predictive technology, which on the surface has created interesting conversations with our research and PM teams, but it, is a challenge to truly add a predictive layer to that data that would create consistent alpha,” Stephenson said. “While that work provides some context around past results, there are many variables that need to be considered on any given trade order that may prohibit full-blown automation or reliance on that prediction. We continue to do more work in this area since there will always be some human biases that can be better optimized over time.”
Another area that Franklin Templeton is investigating is around the ‘wisdom of crowds’ and the benefits of social learning among traders or other investment personnel that can add alpha to its process.
“We think that our huge internal network of investment professionals can be a significant driver of performance if we can harness the power of our collective thoughts to drive decision making” said Stephenson. “We have some pretty creative ideas on how we can systematize those thoughts and dynamically integrate it into our process.”
Since significant alpha will never be created from a ‘black box’, the human judgment component of an active investment process will always need to be utilized as firms “alphatize” technology. “There is a lot of value in managing data, ideas, and processes through technology, and with the right implementation, there is a huge potential for investment professionals to consistently outperform in the long term,” said Stephenson.
The good news is that the results are measurable. On average, institutions who are utilizing Portware’s Alpha Vision are saving between 17 and 22 basis points of alpha capture. “That’s meaningful when you’re managing $200 billion to $300 billion,” Eskandar said. “I believe that further automation into the entire trading lifecycle will occur. Automation has been kept at the trading desk level only, but what we are seeing is that it’s moving upstream from the portfolio manager up to the trading desk and out to the market seamlessly.”
New algorithms use quantitative approach to execution.
The volume grew by 41% in the second quarter of 2019.
The JavaFX-based interface now runs across a greater number of platforms.
The new offering sports 11 liquidity-seeking algorithms with more to come.
GMAG says vacation season is a good time for design, implementation and testing of machine-based trading.