Brute Force Optimizer and Sort by OOS Result

Popov wrote:

Hello Yonkuro,

> Could you please consider to add brute force optimizaton feature

Brute-force optimization is possible, but I'm not sure how useful it may be.

Yes Popov, I think brute force optimizer should come with Monte Carlo and Multi Market Validation tool

So for example I've finished running the reactor, and got the strategy I like, or I already have a completed strategy  and I want to improve this strategy further.

Say I want to find all of the possible combinations parameter of this strategy that make profit on several pairs with tighter acceptance criteria on both Optimizer and Multi Market Validation, and I want the Optimizer to not touch certain parameter such as shift for MA or deviation for Bollinger Band, or maybe SL & TP (which I can untick in optimizer)

I'll get the result I want faster with brute force optimizer than a normal generator/reactor.

Brute Force Optimizer and Sort by OOS Result

Hello Yonkuro,

> Could you please consider to add brute force optimizaton feature

Brute-force optimization is possible, but I'm not sure how useful it may be.


> “Sort by OOS result” will be very useful

Yes, you are correct.
There are things I've not figured out yet. I'll return to that problem later.

Brute Force Optimizer and Sort by OOS Result

Dear Popov,

Could you please consider to add brute force optimizaton feature where the optimizer will search for all possible combinations for a single strategy, just like meta trader optimizer with “Genetic Algorithm” unticked?

This feature should be separated from reactor tab, because it’ll take a long time for optimization process so it’s designed for a completed single strategy only.

Combined with EA Studio advanced “acceptance criteria”, I think it will give a lot of flexibility for end users to create/modify strategy the way they want.

And also “Sort by OOS result” will be very useful, I think you already had kind of “beta version” here https://forexsb.com/forum/post/66760/#p66760

Best Regards.

what does this mean

This is a perfect example of overoptimization (curve-fitting).

It means the strategy takes advantage of particular movements of the market for the "training" period. However, the strategy cannot utilize such cases for the "trading" period.

Curve-fitting happens when we use the Optimizer, or leave the Generator to work a long time. Using generated Take Profit also contributes to curve-fitting.

System Quality Number

Blaiserboy wrote:

For example, I have a scalping system that may endure close to 90 small losses before it gains..... the numbers are atrocious, but the profits are lovely.

I tend to look at the equity curve as the main determinant.

here is a system with some terrible looking numbers, except that the equity curve is lovely as are the profits

I know this post is quite old but those numbers don't really look that bad when you take into account the equity drawdown, clearly your stop loss is very tight and if the trade doesn't fire off like a rocket immediately the trade is nullified and stopped out.

Anyways that's not really what my main question is on this whole thread. I generally try to optimize a strategy which I think will work well with wide parameters using SQN as the "search best" then when it finishes I generally use something like +-10 with net balance to further optimize it in hopes it'll find more trades at the expense of W/L ratio, possibly drawdown and in the end be more profitable over time. The thing is how do I know if my strategy is over-optimized or curve fitted? I know I could use the %OOS to see how those inputs work and obviously forward testing in real time. But is there a way to help avoid over optimizing strategies?

Tweezer A Database Error

Thank you for the report!

I managed to fix the current problem.

It should work well now.

Please report any issues you may encounter!

Multicore cpu and portfolio

Your browser will die way before running 50 Generators.

Usually, 5-6 browser instances will be enough to calculate the necessary strategies.

You can expand the number of strategies per collection to 300 from the settings.

Please "Download" your collections before closing the browser tab. The browser stores all data in a single place and overrides the older data when using multiple instances.

If you want to fill better, run one FSB Pro instance to make the rest CPU cores busy. It will also help to warm the planet faster.

Multicore cpu and portfolio

I have a 32-core processor.

With FSB Pro I can use 100% of the processor. With a browser tab only about 2%.

This means that if I want to use 100% of the processor's power for a currency pair, I have to open 50 browser windows.

If I have a collection capacity of 100 in each window and import all strategies via validator, then equal strategies are sorted out and I come to a value of maybe 60-70 strategies, right?

Issues adding XAUUSD in symbols

> Specifications it says contract size "100" so in FSB i should change lot size to 100 instead of 10,000?

Correct!

Issues adding XAUUSD in symbols

Popov wrote:

Your data import and settings are correct.

The problem is that you have Lot size = 10 000.
Taking the gold price at 1800 USD, it means you need 180000 USD to open one lot with a leverage of 1/100.

Probably you have to reduce the Lot size to 1.

Ahhhh ok, now it shows entry positions
Thank you very much sir! I must thank you for such a wonderful program. I can't imagine the countless hours that went into it.

Quick Question: When i go into MT4 and right click on XAUUSD > Specifications it says contract size "100" so in FSB i should change lot size to 100 instead of 10,000? or 100 refers to 0.01 lots(Micro lot)?