Genetic Optimization algorithm

Dear Mark.

 

When I choose Optimization method in Parameters tab of Optimization, I can set Brute Force and Genetic Optimization.

First I noticed that some time ago, maybe a month or more, Genetic Optimization was working differently, not as good as now.

Before it was slightly decreasing number of simulations, now it is usually set at 15000 tests (unless there are very few tests).

 

I can think of a couple ways to do it on the spot. First is just choose random combination of parameters, sort of Monte Carlo sampling.

Second one is to do some a little bit smart optimization – optimize 1-3 parameters, then the next ones, and so on. And do this a few times in a cycle,

so that the point would converge to the optimal one.

 

Could you give us a general idea on what kind of method are you using for Genetic Optimization? Maybe something similar to what I described, or something different?

 

One more question. What is the difference between Walk-Forward type Exact and Simulated?

Ah, right, it is the same as Brute Force vs. Genetic Optimization. The same question about the algorithm for this one as well.

Genetic optimization is made using genetic algorithms. 

 

They work in a way that it creates a random population of different parameter combinations and then evolves it in subsequent generations to find the “best” set of parameters.

Genetic optimization should be used when there are too many combinations, and where brute force method cannot be used.

 

But it is something else than Exact and Simulated walk forward. Simulated WF is a faster mode that doesn’t run full complete walk forward analysis for every WF combination, but runs the genetic optimization only once and then “simulates” the results for different WF combinations.

 

It is much faster but precise enough to be used normally instead of Exact method.