Beyond a strategic advantage

Combining street smarts with machine learning

Our technological approach ensures system fidelity to superior performance metrics over the long term

  • We develop AI based learning algorithms that adapt very efficiently to changing market conditions. The vast increase in available and discrete datasets coupled with the increase in computing power has enabled massive inroads to be made in the field of financial quantitative analysis. Price action has never been the function of just a few parameters. On the contrary, it is driven by a plethora of diverse inputs.

  • Finding the major price influencers involves the comprehension of the ingredients of price moves and order flows. The simplest answer is supply and demand, but underlying these forces, focus needs to placed on macroeconomic and microeconomic issues, technical factors and other fundamental issues. Each of these areas contains many avenues where past and real time data can aid in future decision making.

  • Moving beyond the traditional
    mind-set of basic technical and fundamental analysis there is a strong case that behavioural science has a role to play in predicting markets. There is also clear evidence of fractal based similarities in price data and many price movements can be related to disparate areas of science such as sound waves, filters and fluid dynamics. With data and computing power thousands of new research avenues have opened.

  • The need for a measured approach is critical. While ‘new-data’ research has been beneficial in many areas it greatly accentuates the risks of model fitting. That is why it is essential to fully evaluate any new technique not just in terms of its performance benefits but also in terms of logical compatibility to the underlying market. While in the search for alpha downside risk still remains the controlling factor.