Pattern-Finding EA | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Intresting finding or simply statistic: 5min candel 14 replies

Price pattern scanner where I choose the pattern 4 replies

Need help finding a scalping broker 12 replies

Need Help Finding a Book 4 replies

Interesting finding: EUR/JPY 1 reply

Attachments: Pattern-Finding EA
Exit Attachments

Pattern-Finding EA

  • #41
  • Oct 31, 2012 6:39am Oct 31, 2012 6:39am
Quoting etrijuni
Hi RR,

Nice EA.
I have back-tested it, the entry point seems good. However, I have difficulty to activate the trailing stop in back-test mode. Any suggestion?
Can't help you if I don't know the settings you used.
  • #42
  • Oct 31, 2012 8:00am Oct 31, 2012 8:00am
Quoting etrijuni
Hi RR,

Nice EA.
I have back-tested it, the entry point seems good. However, I have difficulty to activate the trailing stop in back-test mode. Any suggestion?
The backtest is a little flawed, as it typically does not load all the bars available from the first bar. (well I was finding that)
  • #43
  • Oct 31, 2012 8:34am Oct 31, 2012 8:34am
Good evening, prompt please, what parameters should be changed in a code what to establish the SL of TP&
  • #44
  • Oct 31, 2012 11:04am Oct 31, 2012 11:04am
Quoting Ronald Raygun
Thank you for your indicator and improvements @Paradox.

I went though the code and am wondering how you apply the negative correlation when projecting the future bars onto the chart. It looks like you are combining the results of high negatively correlating patterns on top of high positively correlating patterns,

If you reverse the predictions of the negatively correlating patterns, then you might find similarities between the highly negative and highly positive correlating sequences.

I admit the current trade entry model is flawed in...
Yes I did reverse the prediction on the negatively correlated patterns. I was thinking of mapping it to a future heatmap. But I have been down this prediction road before and there wasn't enough there. I doubt if I have any new insights. I am sure it is probably possible to make this work on higher timeframes though.

What I WAS interested in is, trading itself, post pattern. Basically what if you take any unbiased strategy and you start it right after the identified patterns. Then you only trade the strategy when it matches a pattern that has been successful for it in the past.
  • #45
  • Oct 31, 2012 2:41pm Oct 31, 2012 2:41pm
Quoting deno
I opened a new demo yesterday .
put the ea on 15 H1 charts and it took 10 trades (and 1 this morning)
this morning 7 winners !!! just an accident ? we will see

RR what a great work so far
Wow... Nice work.

Can you update us with the results?
  • #46
  • Nov 5, 2012 8:39pm Nov 5, 2012 8:39pm
Hi, I'm absolutely new to trading and to Metatrader.
I was trying to install and use the correlator in the first page, but, while I'm able to compile it and I see it in the "custom indicators" list, trying o add it to the chart nothing happen.
I"m able to add the Paradox posted in post 38
I've copied the .mq4 files in the expertsindicators subfolder, then I've opened them and compiled them (obtaining the ex4 files in the same folder): is there something wrong/something else I should have done?
I've tried both the original version in the first post and the newer version a few posts ahead. I'm using MetaTrader 4.00 build 438...

To be honest, I don't even know what the difference between the two versions should be is.

Thank you.
  • #47
  • Nov 6, 2012 8:32am Nov 6, 2012 8:32am
Quoting JosefM
Hi, I'm absolutely new to trading and to Metatrader.
I was trying to install and use the correlator in the first page, but, while I'm able to compile it and I see it in the "custom indicators" list, trying o add it to the chart nothing happen.
I"m able to add the Paradox posted in post 38
I've copied the .mq4 files in the expertsindicators subfolder, then I've opened them and compiled them (obtaining the ex4 files in the same folder): is there something wrong/something...

The original file is an EA and not an indicator. You would have to put it in the root of experts folder and not in the indicator folder. You can then run it in strategy tester or real time on your chart and it will make predicted trades.
  • #48
  • Nov 6, 2012 9:34am Nov 6, 2012 9:34am
Quoting Paradox
The original file is an EA and not an indicator. You would have to put it in the root of experts folder and not in the indicator folder. You can then run it in strategy tester or real time on your chart and it will make predicted trades.
Thank you very much!
  • #49
  • Nov 6, 2012 10:25am Nov 6, 2012 10:25am
I haven't placed any trades off of this. I have just been watching but has anyone seen some good results. I also looked into the WmiFor30 that is mentioned in post #35. I think these two ideas are outstanding and I want to stir up some conversation about it.
  • #50
  • Nov 6, 2012 10:54am Nov 6, 2012 10:54am
Quoting fxtrader21
I haven't placed any trades off of this. I have just been watching but has anyone seen some good results. I also looked into the WmiFor30 that is mentioned in post #35. I think these two ideas are outstanding and I want to stir up some conversation about it.
If I revisit this idea I would use these methods instead of correlation.

http://code.google.com/p/jmotif/wiki/SAX
http://cloudeventprocessing.com/2010...-data-cep-sax/

And as I said before I would look to predict when a system would work best and not try to predict the actual market direction.
  • #51
  • Nov 7, 2012 5:41pm Nov 7, 2012 5:41pm
Ronald Raygun,

I've followed some of your works of art for a few years now.

Thank you for sharing this system.

Is is possible to add a time filter? I want to run the EA from the start of the Asian session to the end of London (2-3 hours into NY session). The price action in afternoon hours of NY session does not move much.

I can manually turn it off and on in MT4, but a time filter would be great.

It is very kind and generous of you to share your creation here, and I'm very appreciative.

fyi.. my demo account is running 16 pairs on 1 hour timeframe, and got 200+ pips running Sunday night through 11:00 EST Monday. Monday-Tuesday night produced bad results (loss pips, and I will no longer run on Monday nights)
Tuesday-Wednesday was good (+90 pips).
These results are based on me closing all trades manually with a script at 11:00 EST.
So far, the best three pairs in pips won are Gbp/Jpy, Eur/Jpy, and Eur/Aud.

I'm no coder, so I'm not sure how I can contribute other than sharing my results.

Regards,
pipguy
  • #52
  • Nov 21, 2012 4:14am Nov 21, 2012 4:14am
Hi All,

Always look forward to new ideas from RR. Running tis latest EA on just 5 pairs for a couple of days and have a few active trades. My first thought is that it would be nice to see the prediction remain on the chart when an order it opened - maybe coloured light grey. Then we can see more readily how the prediction actually panned out.

Another observation is to possibly use the current prediction to manage active trades. Here are 2 positions I have open :

Attached Image (click to enlarge)
Click to Enlarge Name: snap1.gif Size: 15 KB


In this first example the BUY looks good. The prediction is that we will hit TP and before SL.

Attached Image (click to enlarge)
Click to Enlarge Name: snap2.gif Size: 15 KB


In the 2nd example the BUY is going against us and the prediction is now that we may just break even. Seeing this I would adjust TP down to the B/E level and be happy with that.

May be an option called AdaptiveTargets in the settings?

I am trying this initial run with periods = 7 on a H1 chart. I might change the MA - Simple to Smoothed to try and get better matches.

B Rgds

WR.
  • #53
  • Nov 10, 2013 8:52am Nov 10, 2013 8:52am
this was looking good.. what happened?
DON'T TRADE THE DIRECTION, TRADE THE MOVE
  • #54
  • Sep 9, 2020 3:22pm Sep 9, 2020 3:22pm
Quoting Pipomagic
this was looking good.. what happened?
Is anyone using or testing this? it had so much promise
  • #55
  • Sep 10, 2020 8:03am Sep 10, 2020 8:03am
Anybody can guide me what setting need to be put in. I tested it all the day and it seems good result from 4300 dollar to almost 4700 dollar. I need to fix lot size with proper tp and sl
  • #56
  • Sep 10, 2020 8:43am Sep 10, 2020 8:43am
today's result almost 1000 dollar in one day
Attached Image (click to enlarge)
Click to Enlarge Name: Untitled.png Size: 31 KB
  • #57
  • Sep 10, 2020 11:10am Sep 10, 2020 11:10am
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot2.png Size: 107 KB
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 155 KB
  • #58
  • Sep 10, 2020 7:53pm Sep 10, 2020 7:53pm
Quoting moaaz.mirza
Anybody can guide me what setting need to be put in. I tested it all the day and it seems good result from 4300 dollar to almost 4700 dollar. I need to fix lot size with proper tp and sl

Ive been messing around with different settings. been having good success on 15min chart
  • #59
  • Sep 11, 2020 10:19am Sep 11, 2020 10:19am
this system is crazyyyyyy good!!! good work hurry with the full version im willing to pay
  • #60
  • Sep 11, 2020 6:10pm Sep 11, 2020 6:10pm
Quoting jaycash
this system is crazyyyyyy good!!! good work hurry with the full version im willing to pay

it needs to be optimized for specific TFs. you will have very different results with same parameters on a 5m compared to a 4h
0 traders viewing now