Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
That guy whoever you mean is Masterbrain? Where did you get that from? And where did Masterbrain attack anyone on elitetrader if looking at his 4 posts? And what's it got to with this thread and my proper criticism?
Do you have enough money for a a flight ticket to Europe? I again invite you to meet here. No need to study Polish because I don't speak Polish at all. And just in case you don't know Central Europe is a small place in comparison to America. A few miles from here to there is like from one border to another.
Ok guys PLEASE lets get on to the point.... Its a very serious issue being talked about in this thread.
If we cannot find a good backtester, or if we discover that the major products (for me thats NINJA, MC, and AMIBROKER) -- and yes I own all 3... Are all flawed.. what do we do?
i was really hoping to optimize and walk forward test to come up with some good automated trade plans.. Is this feasible?
Perhaps I need to rethink all of this, and then realize that perhaps scans are the way to go, to find potential setups and then go in for discretionary trading.
Anyway, its a sad realization that there are serious flaws in software that cost quite a lot of money.
Can we not argue about things, and please get to the point, again, its really serious, as I am putting in a boatload of hours learning how to code systems and if I cant test them properly it sounds like a futile exercise...
Please Mike, I humbly request you keep this thread alive so we can explore this. I know you are having an argument or something with another poster - (yes i understand you must protect your site and have it held to the highest level of integrity). I am just worried this bashing will cause you to close this important thread, and I hope you wont.
I really think futures.io (formerly BMT) is a treasure and am so happy it exists. I hope that this very important thread can shed some light on this very critical problem.
There are serious flaws with almost every software solution i have seen. Or atleast problomatic things that must require work around.
Like i said before. if you do your walk forward optimization or testing, MANUALLY. moving the window with the date setting in the instrument panel. You get reliable and repeat able results in Multicharts. If you rely solely on the automatic walk forward optimizer. yes I believe that they are miss representing the robustness of their platform. and yes it is a major selling point which i think they are basically cheating people on until it becomes 100% reliable. Then your going to have some windows that do not give you results you can repeat using coded date/time flags or the date settings in the instrument panel.
You can verify each window of the WFO by using the time/date panel, adding in the max bars back setting to compensate for lead up, and using the settings given by the optimizer. Some windows will verify results 100% some will not. But the ones that don't are relatively close, 1 trade off using this method. I believe its from trades crossing windows of the optimizer and how they are dealt with. But never was 100% cleared up by tech support or the "engineers" at MC. Also coded time and date flags do not work well at all. Which is a big hunk of BS if you ask me. Since they work 100% in tradestation which is what MC is based on.
I haven't found it to be so troublesome to give up automation. That is not what this thread is about. Actually Its more about understanding the platform your using, its limitations and knowing your results well enough that you believe in them. Not just relying on what a platform spits out when you press optimize or run. I still use multicharts and am still developing strategies on it.
Every platform has its querks. I have recently been making suggestions to NinjaTrader's forum how to fix their WFO open trade boundary handling. Which may or may not be an issue in MC WFO too. Simply NJ WFO exits any open trade on the boundary bar.
This approach makes NJ WFO become part of the strategy logic, which means the NJ WFO is absolutely worthless at best and dangerous at worst.
It looks like MC is having problems. Not sure if they have been resolved since then.
But what about TS WF, is it ok, no flaws? Any feedback is appreciated.
@treydog999 - can you comment on any solutions you have found to date?
I'm interested in the boundary bar issue. As I understand this concept as it applies to WFO it would force an exit on the boundary bar of each OOS period. I'm now using MultiCharts and am interested if anyone has any insight as to how MC handles the bound bar issue? Obviously I'm trying to get a more accurate representation with WFO as many trades span one walk forward period to the next. So if an exit is forced at the boundary I would expect my results to be inaccurate.
Any help with understanding this would be greatly appreciated. I'm also interested to learn from users of NT or Amibroker how those platforms are handling this issue. Additionally, if MC is forcing and exit is there a work around that can be be hard coded that addresses this?