When looking back at the market, ten years ago algo-trading barely existed, but according to research done by a Boston-based consulting firm called the Aite group, one third of all trading decisions in the U.S. markets are now made by machines. The Aite group predicts that by 2010 more than half of trading decisions will be made this way. One really interesting statistic that I mentioned in class was that Deutsche Bank in London has over 70 percent of spot trades carried out each day without any human intervention. The impact of algo-trading is that even if an individual does not own any shares, more than likely shares within your pension fund are being bought and sold using algo-trading.
Algo-trading is an efficient practice considering that computers can make multiple trades, monitor thousands of stocks and do it all at a remarkable speed. Another point is that transactions made by machines can also be done without anyone even noticing! Considering that the market is also reacting to purchases and sales of one another, there is significant profit potential when transactions are not noticed. When a person sees a stock change price, they might react in a few hundred milliseconds. A computerized trader is at least 10 times faster! These few hundred milliseconds might seem trivial, but if the price changes by a fraction of a percent in the split-second before a trade worth several millions, it could mean a difference of tens of thousands of dollars. Ideally a computerized trader offers a trader low latency market access, which means that there is little delay between placing an order and seeing it fulfilled.
Another advantage to an increase in computerized trading is the use of Stealth- trading. Many of the leading brokerage firms now have computers running volume-weighted average price (VWAP) algorithms. The algorithms break up big transactions into significantly smaller and separate transactions so that the rest of the market does not detect what a company may be doing. In an attempt to diffuse the use of these algorithms, Pattern-Spot software works to detect algos trying to sneak in transactions. This type of software also attempts to determine what is going on with these algos by studying the size of transactions.
The final point of this article which I found to be the most interesting were recommendations on how to beat algo-trading in markets. One of the most important concepts is to go with your gut feeling by making predictions in regards to where the market will be going. This of course means using some computerized data to help distinguish additional data needed. Another strategy suggested is looking for signs of liquidity. This means looking for the presence of buyers and sellers for specific stocks can open up a chance for a trade ahead of algo-trading. Unfortunately, signs of liquidity have also opened the door for new types of algo-trading called “sniper” algos which wait for a suitable buyer or seller to emerge and then conduct the trade as fast as possible, before the price can be affected by other traders.
After reading this article, it amazed me to see how much of an impact computerized algorithms have on the stock market. However, one important thing to realize is that human traders can make up for lack of data with instinct and experience. These components are what essentially designed these type of programs in the first place.
Matthews, Robert. "Where have all the traders gone?." New Scientist 194,260602 June 2007 42-45. 13 Noc 2007
7 comments:
For some reason my web address did not show up for my citation. Here it is
http://web.ebscohost.com.proxyau.wrlc.org/ehost/detail?vid=22&hid=112&sid=a54aeeb5-4338-45e4-b43a-38707393d63f%40sessionmgr106
Hey...very insightful article. I was skeptical of the percentage of human involvement in high-volume trading, especially with all the software that has been developed and the algorithms that were mentioned in our trip to Fannie Mae. While highly automated algorithms save a lot of number crunching work, I think it poses a threat to day-to-day investors as it benefits large institutions. However, day-to-day investors use much simpler algorithms in their online brokerage/trading accounts, such as selling a stock when it reaches a certain low price. I'm curious to know how this will play out in an era where data security is one of the largest obstacles for any financial organization to overcome. Thanks for sharing..
Ajai makes a good point. Our class guest Christian H really addressed this point well when he said there is a "gut" element to trading. The bottom line is that good computer programming will not necessarily make your career successful- you need to determine how you will make a difference with your employer.
Very intersting article, algorithimic trading is definately on the leading edge. Did the article mention approximately how many traders are lost due to algorithimic trading and computers?
In training I had at the bank, one of our instructors (former Head ForEx trader at a major firm) said there are 3 keys to trading:
1) have an opinion on what the market is going to do
2) act on it
and 3) be right. By doing two of them, the best you can do is not to lose money. you really need to nail all 3. the computers obviously are pretty efficient at this, but traders will still rely on their own acumen. it's the ultimate risk/reward for the guys on the floor. emotions really run the gamut from guys sitting around relaxed and elated grins to those hunched over sweating staring at their screens. it's hard to believe this emotional side of the business will ever be lost.
It was surprising to know that more than one-third of the trading are conducted without human. As Prof. perednis mentioned, I'm very intrested in approximately how many traders are lost due to algorithimic trading and computers too.
I really found this article very interesting in how trading changed rapidly through time with technology! Like other of my classmates stated: the large about of people engaged in trading through machines.
Post a Comment