Price Action Lab Blog

Premium Market Analysis

Trading Strategies

More on the Next Generation of Trading Systems

I received an email from someone who read my post “HFT is Not the New Generation of Trading Systems“, asking me whether I am “talking about quantum mechanics or something similar”. The short answer is no. The next generation of trading systems will be based on scientific discoveries that are not as widely known as quantum mechanics.

Quantum mechanics is old science and nothing of the short that can mark a breakthrough in the development of next generation of market edges. As a matter of fact, quantum mechanics may help to improve the speed of existing HFT systems by reducing latency. Maybe one or two of the many different interpretations of quantum mechanics can serve as a general foundation for the development of the next generation of trading systems but that is basically it. The details will not be directly involved with established quantum mechanical theory and technology.

Another question  is: will next generation systems be using price (and volume)? The answer is that certainly the aim will be to estimate with high statistical accuracy price direction and possibly the magnitude of price changes but that need not be accomplished by manipulating price data. These systems will be of anticipatory nature and not just reacting based on current and past information. 

We are talking  about something in the early research stages with results expected after at least 5 years.  The details are certainly proprietary but be sure that many are working on their own versions of what a next generation trading system should be like. For example, when I was doing backtesting back in the mid 1980s, I used to say that my trading style was purely discretionary when asked because during that time the ability to backtest offered a significant edge especially for rejecting worthless ideas. Thus, do not expect any serious developer of a next generation trading system to actually come forward and tell you what they after. It would be like burning future profits. Methods of trading last for a maximum of 12 years in my experience. Chart analysis lasted for about that long after it was popularized in the late 1960s. That method worked well in the 1970s. Then, backtesting offered the new edge and it worked well from mid 1980s to mid 1990s.  Then machine generation of trading systems appeared by the end of 1990s. I developed one of the first programs for that purpose as described in my books published in 1999 and 2000 and this approach is still profitable and will be for a few more years. This is how life progresses. Nothing lasts for ever and this is also true for trading systems. If you want to survive you must stay ahead of the crowd.

Comments are Closed

Theme by Anders Norén