Unanswered questions about EA

Hello everyone,

I am an Engineer with background in control theory and computer science.
I have started programing EA few months ago so I am kind of new and I have few questions for experienced developers.

I have developed a strategy by following the different step from http://www.financial-hacker.com/build-better-strategies-part-3-the-development-process/ (by the way this article is so great).

I have now in my hand a strategy that passed all the step described in this article.
By looking at the different trades that the strategy has done, the trades duration vary from 1 week to few weeks (the maximum is 24 weeks). This means that the trade are left open during weekends.

What are the rules for that ? I mean I find a lot of article on internet that warn about keeping the trades open during weekends. It is stated that if there is some bad news, then the price will open with a strong gap and the stop loss limit might suffer from slip.
I don't understand the rationale behind that, I mean the EURUSD currency is stable and it certainly won't fall off from Friday to Monday. What is your advice on it ?
If I need to close the trade during the weekends then the SR of strategy will decrease slightly and it became unprofitable.

Secondly, I am using the WFO analysis like explained in the workshop. Is it something that every strategy should have ?
The market is moving so the strategy need to be adapted somehow but what is the rules ? When do we need to retrain ? During DD or even when the strategy is profitable ?

The rules states is every 35 weeks but I guess it depend that timeframe of your algo...

Last question regards the money management, do you guys are using fixed lot or using a formula for that ?

Thank you by advance !

By the way, I think you guys from Zorro are doing a great work and the manual is perfectly done.
My aim is to also share my strategies !

I always leave trades open over the weekends. Maybe if there were really big news expected (something like Grexit or Brexit) I would stop the strategy but otherwise I think it unnecessarily decreases the profits. But it depends on the money management too...

That said - if the strategy becomes unprofitable when closing the trades over the weekends I would double check it if it really has enough edge.

Some strategies can be run without any optimalization. Sometimes, the optimalization can even decrease the robustness of the strategy. But when you optimize I think you have to decide for yourself what's the best retrain period for the particular strategy. You can run some experiments to determine that.

The EURUSD is not stable, it does not fit the economy of all the participating countries, i.e before Greece joined the Euro they had to pay double digit interests, after that they paid the same 4-5% like other strong countries. It was a nice time for the Greeks to earn money by paying lower interest, but there was a reason why they had to pay higher interests before. There is a high chance that the Euro will break within the next 5 years. If it breaks it will be over the weekend, because people cannot react quickly enough. On the day of the Brexit votum we had a 500 pip movement. This is the minimum you should be prepared if you hold EURUSD positions over the weekend. Just be prepared and use a smaller position.

WFO is good, but at the end you have just one optimal result, you don't know how the neighborhood looks like. I like this https://www.youtube.com/watch?v=B4ScXtiM3OE better. You can find Dave Waltons papers in the internet.

Moneymangement depends on the strategy. Most times I use fixed fractional, but there are strategies that always have a looser after a big winner. Then it is better to use fixed contracts. Other methods use volatiliy for positionsizing. A good book is Van Tharp - Definitive Guide to Position Sizing.

Thank for your response guys.

I am looking a little bit more about position sizing.
In the training, the following money management formula is given:

Margin = 0.5 *OptimalF * Capital * sqrt(1 + ProfitClosed/Capital);

let say that capital=1000$
and with have only 1 asset so OptimalF=1

This mean that the margin is equal to half the capital ? This mean that you are likely to loose half of your capital for one trade ?

I think that is a bit too much. I don't understand how the lot is calculated, it shall be based on the stop limit isn't it ?

About the weekend, is it possible to lower the order size of your trade before the week end ?

Just a vey important question folks, i'm new in here. i have been battling myslf with is question and searching your fourm.

Is it possible to import an MT4 costume indicator because my strategy is based on an MT4 custom indicator and i have its source code ?

In general, yes. You just have to convert the MQL code to lite-C. Usually the lite-C implementation will be much more simple than the MQL version.