Reader Request - Black Box Trading

A reader of my blog, (JJ) who I would like to thank has taken time to read at least one of my posts, brought up a request for me to write about Black Box Trading and Statistical Arbitrage (Stat Arb). Now, I must be fully honest, my level of understanding under these two techniques is limited so I did a lot of research for this blog. I only dealt with a type of Black Box Trading once while working. It was primarily an "iceberg" trade. Black Box Trading, properly known as Algorithmic Trading, is the use of computers to enter buy and sell orders in the equity markets through electronic triggers. For any reader who was interested in starting up Black Box Trading, you may already see the first hurdle. This style of trading is commonly used by hedge and quantitative funds because they manage millions or billions of dollars in assets and have the practicalities of capturing small movements in regular trading. It may be used to enter orders before information has been fully received by human traders or to create liquidity by market makers. It may also be used to "iceberg" orders, a technique used by firms who require to buy or sell a large quantity of shares but want to limit market impact, thus divide the orders into smaller units and prevent volatility. The idea that one program can run a successful Algorithmic Trading style for many years is also false. According to an online post, the systematics of setting it up is enormous. The steps in the link will be costly, which present a second hurdle for start up for retail investors. The author of the post also points out that a regular PC does not have the capabilities to handle these tasks. Algorithmic Trading has also been quite controversial. High-Frequency Trading may require Black Box Trading systems and accounted for almost 75% of volume in the U.S. Markets in 2009. Many blame High-Frequency Trading for the volatility and crashes of 2008, as computer models sold off large quantities of stocks as markets and indices would touch key order entry levels. As these key levels were breached, computers would sell and eventually reach more key order entry levels, creating more selling by computer models. I could continue further with more points in my research about its advantages and disadvantages, but my main point throughout this blog was to inform my readers that Black Box Trading is not realistic for retail investors. JJ also wanted me to write about Statistical Arbitrage, but again, it requires a lot of data mining and is not practical. To quickly summarize Stat Arb, it requires a trader to purchase and short many securities (sometimes hundreds) in the same sector, region, industry, etc. This is done to eliminate beta, which is a numerical value that measures a stock's relative return against the market (A beta of 1 means the stock moves almost parallel to the market. A beta of 2 means the stock moves double the market. Negative beta occurs when a stock typically moves in the opposite direction), and other risks. Once this is done, the trader may go long the basket that is under performing and short the basket that is over performing. The theory is that stocks under performing will catch up to the rest of the market, and vice versa, and hopes that the profit on one side is better than the loss on the other side.