07.21.2016
By Rob Daly

Robo-Advisors Eye Active Management

Robo-advisors are no longer just an automated tool for “set it and forget it” investors looking for passive investment strategies.

Trading-tool provider Trade Ideas has launched its AI-based robo-advisor, dubbed Holly, that locates and exploits market opportunities for investors, according to David Aferiat, a co-founder and managing partner at Trade Ideas.

“Unlike passive models that incorporate the entire portfolio, our model is to be applied to a fraction of the portfolio, where investors would like to see some outside returns rather than indexing themselves to an average performance,” he said.

The Holly engine scans the previous day’s market data and first applies 40 to 60 base strategies to the universe of stocks.

“If one strategy was to look at 52-week highs, Holly looks through our library of indications to see if a 52-week high with more than average trading volume performed better than a 52-week high with less than average trading volume,” explained Aferiat. “Then it continues in an iterative process to add following variables like social media mentions until it has gone through all the indicators that we have.”

Holly then collects the optimized strategies and automatically back-tests them to suggest the best performing strategies. The process involves millions of calculations that take place on Trade Ideas’ internal data centers.

“This is a big data exercise where we orchestrate all of our serves to do specific functions collectively,” he added.

After going through this “quantitative combine,” Holly produces a number of trading strategies based on the potential market opportunities. “Holly produced 10 strategies today and only five yesterday,” he said.

Holly then scores the final strategies against 12 criteria that provide them with a winning percentage and probability of success. “These strategies can involve holds for as little as three hours to as long as three days.” said Aferiat.

Once the market closes, the process begins again.

More on Artificial Intelligence:

Related articles

  1. With Eugene Kanevsky, James Redbourn, and Joanna Wong, CLSA

  2. AI/ML on the buy-side trading desk is a long-term program with short- and mid-term deliverables.

  3. Aiden uses the computational power of deep reinforcement learning to improve trading results and insights.

  4. OPINION: Artificial, Yes. Intelligent? Maybe.

    BondDroid’s AI-generated prices are integrated directly into LTX’s pre-trade analytical tools.

  5. Fintech firm has partnered with Jim Toffey, founder of Tradeweb Markets, to create LTX.