When I was first exposed to the concept of quantitative value strategy and was really excited about it. Recently, I realised that such strategy is already prominently practiced in the market under the name of smart beta. Smart beta strategies basically aim to outperform the capitalization-weighted market based on factors such as value, size, momentum, and volatility. The strategies boasts superior and consistent returns based on positive back testing results on historical data.
Then there is smart alpha, a strategy that is ever evolving and proprietary to the creator. While smart beta is backward looking, transparent, methodological and predictable, smart alpha is forward looking, versatile, dynamic, and evolutionary.
I was intrigued by the possibility of building a system that can tell me when to buy or sell what investments. All I need to do is to let the system do its job while I can truly live my life.
The concept is similar to creating an Artificial Intelligence not unlike a Chess AI. For every move you make, the AI will extrapolate every single possible moves to achieve the ultimate outcome of checkmating your king.
Your "chess moves" will be akin to the movements in the market such as changes in interest rates, or the flow of money in/out of various asset classes etc. The outcome of "Checkmate" will be akin to generating the highest return possible at the highest probability rate trade off.
The best part of this AI is its ability to learn from both historical and future events and adjusts its calculations of probability and re-calibrate its strategy accordingly.
AI that can think and learn is no longer something in a science fiction but very real in our world today. There is only so much a human brain can process and we are also highly susceptible to our own biases that we allow emotions to rules over logic.
I believe this is a worthwhile pursuit and would be devoting a part of my time towards this end goal.
No comments:
Post a Comment