Holy-Grail!! At-Last - page 4
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Have a look at this article - https://www.mta.org/eweb/docs/2007DowAward.pdf - it talks about volume weighted MA and the volume price confirmation/contradiction that may be gleaned from it. If you want the source code for VWMA and VPC indicator, you can get it here - https://www.mql5.com/en/forum/126886
Wow..... this has really got some potential. Your version of the Vwma didn't work for me. However, I've managed to download a Vwma indicator from http://russian-forex-indicators.blogspot.com/2009/06/indicator-vwma.html. Let me know if it's doing the same thing as your version. The very first test, though short from 1-1-2010 to 7-1-2010 shows a positive result without any optimization. I used your Period of 10, Median Price for Vwma and SL & TP of 50. I always use 50 to begin with. The pictures are below. I was impressed that it stayed flat giving the number of order it was shooting out. Anyways, next step is to optimize and see if that unlocks any special powers :)
Sigh :( Optimizing over 10 years didn't produce any positive results. Parameters optimized where Period,SL & TP. I'm gonna try increasing the parameter length but I doubt it's gonna produce any viable results over 10 years. Humm... lets try re-writing the rules and have it buy or sell only when prices crosses the Vwma and see what happens.
After adding the above rules and some optimization, here's what we've got after 10 years.
Now we're gonna run this baby with those parameter and see what our curve looks like hehe. Humm.. unacceptable as it's going about 4 years at break even before making that nice win in the middle there. So, lets see, maybe if I add my logical close, force reverse and trailing stop functions to it, that might help. Logical close means if a buy-trigger comes in while its selling, it'll quit the sell and take the buy and-vice-versa. Force reverse means if the last order was a buy, I'll force the next one to be a sell. And trailing stop is trailing stop, don't think it's gonna help much here as the initial stop is large to begin with but I'll give it a shot as well.
Well, I threw everything in my arsenal at it, and nothing else made a significant difference. We could go on adding filter after filter like Hour() or Volume() greater than <no wait it's a volume indicator> but that's just adding more curve fit. One thing for sure is the more filters you add the less trades it's going to place and the less chance of getting any statistically valid # of trades (if you believe in that sort of thing). Now compare that with Macd below and that is why I discount volume based indicators.
Lol .. someone reborn my old grail. Funny seeing this again.
Huh, well it was on the first page, so i assumed it was you ;)
Now that it is alive again: did you make progress, or abandoned the system?
Huh, well it was on the first page, so i assumed it was you ;)
Now that it is alive again: did you make progress, or abandoned the system?
Somehow, I let this system go because I wanted something more powerful. Must have been right after I'd seen your Double account in one week real results. Then sometime later like about 6-months, I re-tested the system and it fell flat for the previous 6-months period. At that point I drew the conclusion that static stop-loss systems could not generate the types of consistency which I was looking for. Since that time I've been moving further and further away from stop-loss systems and closer to grid like systems.
The 7bit Snowball thread was an inspiration toward this new path. One theme which kept repeating itself was that grid like systems tend to produce results within forward/live testing which is consistent with the average returns per day/week/month etc to what you had seen within its back-testing statistics. Also, these types of systems tend to survive blind back-tests longer before they crash. By blind back-tests, I mean back-tests without optimization.
Just looking at your current Results, it should be fully clear that a Non-Grid, Non-Multi-Currency, Non-Hedging strategy cannot pull off those types of returns within 7-days. Where the Grid trader may go wrong is assuming that somehow, he/she does not have a Risk-Of-Ruin and sit on the account thinking its going to make a billion within a year. This is where I think we have to gage success a little differently and use some common sense.
You've just double the account within a week. This is not a retirement account rather a trading account. Take your initial investment out and let history repeat itself again. If you've already gathered your testing stats correctly then you know that the market would only knock a system like this to the ground once or twice within a year if the market's lucky. Why have all the profits you've generated within the account when this freak show happens.
So anyways... is that a real account? If you rather us talk about some of these advance techs in private then send me a pm.
xxxxxxxxx