Sleytus, may you give me email address? I'm sending you the files. Thanks

I have download and install twice,but it will still auto deleted/uninstall from my computer??
Please advice,thx

Haha,alright noted.
I will try to check my antivirus system and see how it work..

Thanks, Popov.  Yes -- I agree it is probably a signing problem.  I also use the Inno Setup installer.

Thanks for the link to the digital certificate for Windows applications -- I'll look into it.  As a side note, we also develop software for the MacOS and it is even a *bigger* pain.

Thanks for sending the files.  To summarize -- everything works, but I did run into some problems that are related to mixing lot sizes with percentages.  Two important points:
(1) A portfolio can NOT be a mixture of strategies that use lots and percentages -- choose one.
(2) If using lots, the 4 input values should be in terms of lots.  If using percentages, the 4 input values should be percentages.

The problem I ran into was that I only changed 'Maximum Amount' to percent but left 'Entry Amount' as 0.01.  When using percentages the strategy code treats that value as a percentage -- and 0.01 is 100x too small.

I've modified Portfolio Maker's UI in a way to make this less confusing.  It now forces you to choose between Lots or Percentages, and depending on your choice will use appropriate defaults for the 4 input values.

I've updated Portfolio Maker -- so please give it a try and let me know.  The download link and license key remain the same:

Download:
http://nutsaboutnets-downloads.s3.amazonaws.com/PortfolioMaker_Setup.exe

License Key:
PORTFOLIOMAKER-BETA-20171101

D5QM54S7 -- earlier you reported a build problem, but I couldn't reproduce it.  I did notice your strategies use quite a number of "exotic" indicators, so I did have to load those into my FSB-Pro installation in order to create *.mq4 files for your strategies.  Also, your strategies did not use magic numbers for their filenames, which Portfolio Maker requires.

Given the *.xml strategy files you sent, below are the steps I followed to get everything to work.  Perhaps others would benefit from this as well:
1. Downloaded the 3 *.xml strategy files that were sent to me.
2. Renamed the *.xml strategy files as magic numbers.
3. When attempting to open the *.xml files in FSB-Pro then FSB-Pro complained about missing indicators.  So, I loaded the missing indicators from the FSB repository.
4. After loading the *.xml files I then exported the corresponding *.mq4 files.
5. Launched Portfolio Maker and 'Added' the *.mq4 strategy files
6. Adjusted the input values to use percentages.
7. Pressed the 'Create' button.
8. Launched "MetaEditor' and compiled the top-level *.mq4 portfolio EA file (i.e. the one you named aaa.mq4)
9. Browsed the build log to confirm the portfolio EA built cleanly.
10. Used MT4's Strategy Tester to run and test the new portfolio EA.

I've added a new feature to Portfolio Maker I wanted to share.  Currently it is only a "proof-of-concept" and not yet available.

The way I've been testing strategies is to create a portfolio EA with 50-150 strategies and run them in a live micro-account for a couple of weeks and then use MT4 Tracker to separate the winners from the losers.  Sometimes the portfolio EA is a net winner, sometimes a net loser -- since it has a mixture of good and bad performers.

The new feature implements a very simple idea -- I set a threshold WinRatio as an input value (e.g. 0.60) and keep track of each strategy's most recent 10 trades.  If the WinRatio for a particular strategy drops below the threshold then it is disabled and no longer able to open a new trade.

Using MT4's back tester I ran the same portfolio EA twice under the exact same conditions -- except the first one did not use the WinRatio filtering, whereas the second one did.  You can see in the second case the total number of trades is much less, yet the overall net profit is the same.  This was a quick and dirty test run on one portfolio EA that I chose.  Results may differ with different portfolios -- I don't know. 

I think the idea makes sense.  What I have found in my experience is there is only so much effort I can put into creating strategies with great statistics -- in the end some strategies perform well and others do not.  Perhaps a better approach is to incorporate some empirical (i.e. performance) rules -- that is, while strategies are actually running then apply simple filters that turn them on/off based on their performance.

When Popov introduced this new concept of a portfolio EA it opened new doors -- not only from a trader's perspective but also in terms of what could be done in software.  In a portfolio EA all the strategies are now running within the same context of code and, therefore, simple policies can be applied that affect some or all the strategies.  This could not be done if each strategy was running in its own MT4 chart.

Auto-filtering is great step!
Will test it after ur implementation of course

I'm not familiar with those systems -- or that terminology.  I suspect you are referring to systems that can tolerate a low WinRatio since their reward is high.  Perhaps a different type of filter would work or dividing your strategies into two, separate portfolios where WinRatio makes sense for one of them.

Technically, since all the strategies live within the same executable, then any type of filter you could imagine can be applied.  Whether the filter makes sense or how difficult it is to code, is a different issue.

I'm not sure that % drawdown makes sense.  % drawdown computes the delta between the previous high point and the current low point.  But in a portfolio with many strategies, there may be several strategies that contributed to the low point.  So, which strategy gets the blame?  Does the last strategy get blamed for the drawdown even though others contributed to it?

To answer your questions:

1. The program is coded to allow up to 1000 strategies -- the most I've used so far is 130.  MT4 is rather CPU-intensive, so this might differ from machine-to-machine.  Also, MT4 is designed for each chart to run in its own thread.  If you have multiple charts open and certain ones require a lot more computation time than others, then there may be some unintended side-effects.  I've Googled, but have not found any more information on this.  More importantly, the problem I've run into is I am often surprised by the number of trades that get opened.  Since I'll use live micro-accounts, then sometimes it gets uncomfortable because a lot more trades get opened than what I had gotten use to when trading only stand-alone EAs in their own chart windows.  The bottom line -- it will require some experimentation to determine how many strategies to include in a single portfolio EA and how many portfolio EAs you can run within a single instance of MT4.

2. I don't know what the optimum number would be -- that would will require some experimentation.

3. In an updated version of Portfolio Maker, you can choose 'Lots' or 'Percentages'.
    a. The input values from Portfolio Maker are applied globally and will overwrite the strategy's original values.
    b. Yes -- Add, Winner and Reduce should all work as before.  Portfolio Maker doesn't alter Popov's code or logic -- it just
        rearranges the code so multiple strategies can be combined into one.
4. Correct.  The portfolio EA can NOT contain a mixture of symbols and time frames.  If you attach the portfolio EA to a EURUSUD H4 chart, then that is what is going to be traded.
5. Yes -- you can continue to use MT4 Tracker.
6. Technically, one license key per machine.  However, the licensing system I use permits some "wiggle room" and will allow 3 activations before the license server in the clouds begins to complain.  If you need more than 3 it is not a problem -- just let me know and I will send another.  The purpose of the license key is to protect the software -- not to keep charging someone who has already purchased the software.  The Beta version will be active for another week or so, and then I will release an official version that will require a license key.  For people who participate in this forum -- if they are interested -- they can send a PM (private message) and I will send a license key at no charge.  Again, this won't take affect for another week or so.

Hi Steve,

Just to keep you posted.

I did a "Stress" test yesterday.  I tried to "load" the MT4 with as many EA (1412 EA in each Broker ) with 2 different Brokers.

In 1/2 day's of trading period, I noticed that in one broker, there were a total of 1,279 trades being transacted and 330 EA being activated

And another Broker has a total of 8631 trades and 531 EA being activated (in the same time period for these 2 Brokers with the same set of EA).

It seem like there are some Broker's server can't handle too many transactions or because of these Brokers place a limit in the total number of open positions at a given time to prevent overloading of the servers if there isn't any limit on all clients.

The Second Broker with high number of trades is from a more well established company and thus probably has a bigger server capacity to handle higher number of transactions at any given time.

The 1st Broker has a limit of 200 open transactions at any given time but one need to write in to request for the limit to be increased to 500 max.

Anyway, I think most people wouldn't have the kind of issues I'm facing as my EA tend to have very high frequent trades per day (average 30 to 50 trades per week/per EA or even 100 or more depending of the systems).  But I'm posting it just to highlight the possible issues others may encounter if they were to load each MT4 with many EA.

Hi Steve,

Yes, I do agree with you that it is better to keep it below 100 EA per MT4 instance.  In additional, I'm using the E-trailing to add trailing functions to all open trades and this E-trialing EA can't function properly (can't trail all open trades) when I've too many EA running.

I'm just doing the experiment to see to what extend I can cut down the number of MT4 instances.  If I'm going to keep 100 EA per MT4 instance (which the same in the past), I wonder how much computer capacity I'm "saving" or free up. 

Nevertheless, it is still way faster to set up 100 EA via Portfolio Maker than to manually set up these 100 EA 1 chart at a time.  With my 1400 EA, I only need to set up 14 MT4 EA as compared to 1400 MT4 Charts!!!!