Find a needle in a haystack
This strategy optimization method makes it possible to find workable solutions amongst billions of combinations, whereas exhaustive optimization would take an unacceptably long amount of time. Genetic optimization does not try every possible combination. Instead, it uses complex algorithms that use principles of biological evolution. It randomly picks a sample of inputs and tests for positive results. The best inputs are kept and the worst ones are replaced with new random inputs—survival of the fittest. You can even customize your optimization by changing the number of generations, population size and the convergence type.
Make sure your results are robust
If the algorithm thinks it has found something good, it will start testing around the newly found peak to determine if this peak is an outlier. If this result is an outlier, genetic optimization leaves it and moves on to other areas. In fact, the more possible combinations you have, the better genetic optimization works. After you complete genetic optimization, you can test the inputs you found even more, with exhaustive optimization, walk-forward testing, or technical indicators.
Find new opportunities
Genetic optimization makes it possible to find new opportunities when creating trading systems, as opposed to fine-tuning an already existing system. For example, you can apply many indicators to your chart, each one of which has literally thousands of parameters. Then you can check if these indicators help predict price patterns by including and excluding them by telling genetic optimization to turn them off and on systematically. You can then use these findings to filter out the indicators that don’t work, and keep the ones that do.
Explore genetic optimization further
Genetic Algorithm was invented to mimic some of the processes observed in natural evolution. The idea with GA is to use this power of evolution to solve optimization problems. To fully understand how genetic optimization works, we recommend you read additional professional literature on genetic algorithms. Explore genetic optimization further.