|
Neural networks in computer software herald the beginning of programmed ability to learn from live events as they unfold (in this case price action and risk control). Neural networks can draw on this learning to make more appropriate decisions as events unfold on a real time basis.
The Build Your Trading analysis is largely based on the chaotic feedback model of market price cycle behavior and fractal analysis. There are parallels in the study of fluid dynamics and weather pattern prediction.
The neural networks in the Build Your Trading price analysis study a fractal level length of just over three weeks. This means that on the average the system gives a trading alert about every three and a half weeks or 15 times a year.
|
An approach to neural networks can be described as follows. 'Neurons' are simulated by mathematical algorithms (formulae). Every interconnection between these is represented by its level of importance when a price event occurs. Each 'neuron' evaluates the signals coming into it. If the combined signal exceeds a particular threshold, the 'neuron' fires and sends a signal to algorithms or 'neurons' at a higher level. If the combined signal exceeds certain thresholds, a reversal alert is triggered. There are multiple layers of these 'neurons' which are organised in a particular way. The key to this neural network is that the programmer has caused it to learn to act on certain trains of events.
You will not find the analysis settings in Build Your Trading in any other program or trading text
book. The analysis in the program is copyrighted and is specific to Build Your Trading.
|