Thinking about AI is making my head hurt – need a day off. Rather than try and work through a whole financial model, will be looking at fractals and financial visualization. If you are already very familiar with fractals, complexity, can fly through mathematical and physics equations with ease – Lee Smolin’s Time and Symmetry in models of Economic Markets (I’ve got as far as the end of the first page, so far) may be a more entertaining read. Later today will post about an attempt to make a fractal and if possible using financial data. I have never tried to make a fractal before.
When looking at financial visualization it can be easy, especially when short of time, to view a graph and make assumptions – but when there are multiple indexes which have different variables – it gets complex quickly – this is an example of something we did for a revision lecture at Surrey in March, data
If you look at the data you have rates which are 5.5 and rates which are 3000 or higher – so trying to visualize these is difficult because of the different scale involved – e.g if you compare on the visualization FTSE All, Warner and Great Portland Estate some of the movements look very erratic. In the end we converted rates into percentages but this is still not an accurate representation of all the different fluctuations in the data – it is too varied and too complex. However based on what I have understood, using fractals to visualize this data might be better because it takes account of scale and dimension. The Pictorial Essay – a Fractal Gallery in Misbehaviour of Markets is a brilliant starting place because Benoit Mandelbrot explains each different type of fractal and looks back at the original Mandelbrot set as well.1
The tool I will be using is available at http://www.ultrafractal.com/
1. Mandelbrot B, Hudson R, (2004) The (Mis)behaviour of Markets, A Fractal View of Risk, Ruin and Reward, New Edition, Profile Books