Sell-Side Electronic Trading:
Industry Reacts to New Generation of Hackers | A New Way to Attack Market Data Costs | The Hidden Cost of Trading | In Fixed Income, FIX Is a Post-trade Solution | Algorithms Multiply to Match Trading Strategies |
On the Algorithmic Edge With JP Morgan's Carrie
November 21, 2006
Over the past year, competition has intensified among the biggest Wall Street firms as each tries to break out of the pack with the fastest and most sophisticated algorithms and related trading tools. In this race, leads tend to be fleeting; product announcements take on an air of "can you top this?" One firm that has consistently been in the front ranks is JP Morgan Chase & Co. Its electronic client solutions (ECS) division last month trumpeted AlgoAlert, an instant messaging service that provides unique charts as well as real-time updates of algorithmic trading performance and trading conditions such as halts and oversized orders. AlgoAlert can be received via a variety of e-mail and instant messaging platforms, including AOL's AIM, the Bloomberg Professional messaging system and Google Talk.
"AlgoAlert is a tool that tracks algorithmic trading performance when traders need it the most--while they trade," Carl Carrie, head of product development in ECS, said at the time of the launch. "It gives traders a competitive edge by ensuring they receive information they need for split-second decisionmaking. The result is improved awareness, better execution and improved transparency in algorithmic trading."
AlgoAlert works in conjunction with any execution management or order management platform, including Neovest, JP Morgan's own broker-neutral execution management system. In addition, the firm recently introduced two algorithms--Aqua and Arid--designed to trade anonymously on exchanges and electronic communications networks as well as in "dark" crossing venues.
In an interview with Securities Industry News markets reporter Alexa Jaworski, Carrie highlighted some of JP Morgan's recent algorithmic accomplishments and explained how it intends to keep ahead of the competition. A 23-year financial industry veteran, Carrie was previously North America chief business technologist for equities at JP Morgan. Before joining the New York bank in 2002 he was founder and president of software company TheBeast.com.
What is your current role at JP Morgan? I lead the
creation of new algorithmic, quantitative and execution products
for equities. Managing the delivery of product development is a
business role, and I work closely with our team--a rich mix of
traders, algorithmic technologists and quants who are all
innovators and practitioners at the same time. That is, they can
conceive, build coding specifications or develop parts of the
software themselves. They use and even market the products that are
developed to clients, which, in our view, is the most extreme form
of agile product development--not just agile software development.
This product development team works closely with our partners in
technology to industrialize and implement our applications, and add
security- and compliance-related features.
What are some of the unit's greatest accomplishments over the last year? I believe we have created one of the broadest and deepest product suites on the Street, ranging from single-stock to portfolio algorithms, from pre-trade through post-trade analytics, to automated algorithmic alert bots--automated agents or silicon-based robots programmed to carry out tasks for the trader--all wrapped in a unified client service model that is secure and information-leakage-free. Our algorithmic capabilities are market-leading in two key areas. One is portfolio algorithmic trading with our Trading Algorithmic Optimizer. With this, we have the first automated and transparent platform to trade portfolios that integrates Web 2.0 technology with a true optimizer. Its real-time analytics allow you to go from a point on the efficient frontier to trading using optimized settings for algorithms with a few mouse clicks. We make the math easy for the trader to use and align the trader with the portfolio manager's goals in a visual, information-rich way. The second area is in how we leverage dark pools of liquidity. This targets the extremes of very aggressive trading, where the objective is to maximize alpha by minimizing timing risk and utilize a sophisticated dark and displayed order model, or very illiquid situations such as for many small-cap stocks where the objective is to minimize market impact and spread costs and operate in a stealthy, completely information-leakage-free way. Aqua and Arid, we believe, are the first algorithms that straddle the dark and displayed marketplaces in a unified and quantitatively rigorous way.
