This is a machine learning extension of the Elliott Wave principle.

- Blogs
- Elliottware: machine learning extension of Elliott Wave principle
- Expert advisers and Elliott waves: Elliotware approach

By JohnLast 2947 days ago

Her I just want to mention that in another blogpost I used BPNN family of neural nets to make market predictions. However the same idea apply not only to neural nets but also with the optimization process in Metatrader.

So as the pricniple is the same how that can be implemented in practice.

It would be again a three steps procecc.

1. Identfiy where we can find an Elliott wave structure.

2. Here you do not need to count the number of bars. You just need to write down the dates.

3. You take the dates and you make an optimization in the Metatrader optimizer.

4. You cross the fingers.

There is a lot to be said about optimization. Most of the experts prefer to not use really the genetic optimizer but the normal optimization. In order to do say they limit the number of the parameters to optimize. that is necessary because with the genetic opimizer you can and you will find a local optimum of the solution that is a peac (the genetic optimizer allows to find a solution when you optimize many parameters). When you optimize looking for only two parameters you can see what is really going on. Do you have a peac of the solutions or you have a vast plain of good solutions. In the first case a slight change in the parameters would lead to a completely different result. In the plain of solutions a slight change even a moderate change will not lead to a substantial change of the profitability.

The plain of the good solutions is always a preferred way. So if you have more than two parameters you need to optimize them two by two. And that is a LOT of work.

Here is one of my favorite articles ever.