Robot Lab Project | Page 58 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Financial Markets Observatory Lab 1,095 replies

Indicators Lab 236 replies

Rel's Lab 103 replies

Nikola's Trading Lab 247 replies

Spud's Lab Accident 22 replies

Attachments: Robot Lab Project
Exit Attachments

Robot Lab Project

  • #1,141
  • Mar 16, 2016 12:43pm Mar 16, 2016 12:43pm
Could u answer my questiom, abokwaik ? did u code all ea in this thread or using EA builder?
  • #1,142
  • Mar 16, 2016 3:47pm Mar 16, 2016 3:47pm
Quoting beyeu123
Could u answer my questiom, abokwaik ? did u code all ea in this thread or using EA builder?
Are you serious ?
Using an EA builder of course.
No guts, no glory
  • #1,143
  • Mar 16, 2016 3:51pm Mar 16, 2016 3:51pm
Quoting beyeu123
Could u answer my questiom, abokwaik ? did u code all ea in this thread or using EA builder?


I think I can answer that question for him.
His quality of work is in the top three on this site.
As for EA builder a lot of the coder's will not even touch it.
He is a artist who writes his own code's and is willing to share his work.
That's my two pence worth.!
  • #1,144
  • Mar 16, 2016 5:08pm Mar 16, 2016 5:08pm
Quoting beyeu123
Could u answer my questiom, abokwaik ? did u code all ea in this thread or using EA builder?
EA builder is ok if you will test easy strategy in 10 minutes, but if you will real trade with EA, it is better to type own. Or you can make own EA/indicator builder, for effeciency in creating fully or half automated dashboards - your code is under your control

abokwaik code here was his own
  • #1,145
  • Mar 16, 2016 8:44pm Mar 16, 2016 8:44pm
Quoting abokwaik
{quote} Yes. Its like percentage change. I will build the MT4 version based on that.
Btw, I've done manual GJ/EJ (two pairs that I assumed correlated) check over 1 year, not sure if it is consistent in profit over longer period, pls see following result:
Attached Image


I also have further reading regarding correlation. I found out that "cointegration" approach as trading system is more appropriate than "correlation". Pls check out about cointegration https://www.quantopian.com/posts/how...-on-quantopian .

Interestingly, I calculated over a spreadsheet that GJ/EJ is not cointegrated enough. The p-value is only 0.744248 (the closest to zero the more it is cointegrated).

Here I plot the spread between GJ and EJ over 1 last year (Mar 15 to Feb 16). If it is cointegrated, the spread shall be ranging over a fix range. However, we see a widen spread.

Attached Image


So I would love to suggest you to develop cointegrated trading system instead of correlation one. I think the program should be able to scan over many pairs automatically and recommends a result which pair of pairs is cointegrated.
  • #1,146
  • Mar 16, 2016 9:00pm Mar 16, 2016 9:00pm
Quoting abokwaik
{quote} Here .. . TrioScalper EA EDIT : I think the version I posted was using MACD Signal instead of Histogram. So I deleted that. This one uses MACD histogram as per system description. Sorry for any inconvenience. {file}
Thank you, interesting claim from the thread starter. Let me forward test it on demo.
  • #1,147
  • Edited Mar 17, 2016 12:04am Mar 16, 2016 11:26pm | Edited Mar 17, 2016 12:04am
Quoting gwindy
{quote} Hi Khalil, Sorry to bother you so soon...I have been forward testing and it appears the trades are not opening correctly (as per the rules), please see below chart : {image} The second trade on the chart duplicated a second order (two orders placed at once). EDIT: Signal candle must open below MA and close above MA for valid BUY (and opposite for sell) Thanks again
I am not sure if candle shall cross the MA

The TS said just above the MA for buy, and below the MA for sell.

Maybe if we want to prevent duplicate order, just open trade after the first tick of negative/positif MACD confirmed. So we trade the MACD cross.

Edit:
I got good impression with the system, manual trade on demo. I traded the MACD value crossover to 0.
Attached Image
Attached Image
  • #1,148
  • Edited 5:31am Mar 17, 2016 1:12am | Edited 5:31am
Quoting HeWee
{quote} I am not sure if candle shall cross the MA The TS said just above the MA for buy, and below the MA for sell. Maybe if we want to prevent duplicate order, just open trade after the first tick of negative/positif MACD confirmed. So we trade the MACD cross. Edit: I got good impression with the system, manual trade on demo. I traded the MACD value crossover to 0. {image}{image}
No worries HeWee. The issue or problem I have found is defining 'just above or below'. I am finding that waiting for a signal candle to crosses the MA and close typical filters out a lot of bad signals. It doesn't rely on the MACD histo crossing from negative to positive (vice versa). As long as the MACD histo is a positive I go long, and negative for a short- assuming the other criteria align of course.
  • #1,149
  • Mar 17, 2016 8:39am Mar 17, 2016 8:39am
Hello Abokwaik,
After 20years of trading in futures and forex, I found 1 technical configuration which works very very well. In which one i ghave absolutly no fear to enter a position.
Although it is very simple, it works.

BUY:
Wait for an engulfing reversal candle which closes above 15 or 20EMA. Enter position on pullback of the engulfing (38% to 50% retracement)

SL Below the high of the previous candle before the engulfing.
TP 1/2 at the last swing high and 1/2 to determinate.

Sell: opposite rules.

Would you start to code an EA based on this?

Thank a lot for your fantastic work.

S
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 292 KB
  • #1,150
  • Mar 17, 2016 6:02pm Mar 17, 2016 6:02pm
Quoting sbenichou1
Hello Abokwaik, After 20years of trading in futures and forex, I found 1 technical configuration which works very very well. In which one i ghave absolutly no fear to enter a position. Although it is very simple, it works. BUY: Wait for an engulfing reversal candle which closes above 15 or 20EMA. Enter position on pullback of the engulfing (38% to 50% retracement) SL Below the high of the previous candle before the engulfing. TP 1/2 at the last swing high and 1/2 to determinate. Sell: opposite rules. Would you start to code an EA based on this?...
Why not, I just need to add few modifications on the Engulfer System.
No guts, no glory
  • #1,151
  • Mar 18, 2016 5:34am Mar 18, 2016 5:34am
Hi again Khalil did you conclude the indi center of gravity can be used to put pending orders on the borders of tha tindi.İsnt it
  • #1,152
  • Mar 18, 2016 5:56am Mar 18, 2016 5:56am
Quoting falanca
Hi again Khalil did you conclude the indi center of gravity can be used to put pending orders on the borders of tha tindi.İsnt it
Yes falanca it can be done. Just allow some time as I am really busy these days and can hardly find time for MQL coding.

Thanks
No guts, no glory
  • #1,153
  • Mar 18, 2016 6:36am Mar 18, 2016 6:36am
Quoting abokwaik
{quote} Why not, I just need to add few modifications on the Engulfer System.
exactly this is what I was thinking about.Thank youS
  • #1,154
  • Mar 18, 2016 8:49am Mar 18, 2016 8:49am
Quoting abokwaik
{quote} Yes falanca it can be done. Just allow some time as I am really busy these days and can hardly find time for MQL coding. Thanks
Ok.İ just want to make know you about my conclusion.But if yode this ı will be happy also.Best Regards.
  • #1,155
  • Mar 24, 2016 10:32am Mar 24, 2016 10:32am
İ just want this topic up
  • #1,156
  • Mar 27, 2016 1:57am Mar 27, 2016 1:57am
Quoting abokwaik
Close All Treades EA This is an update that allows Profit inputs to be negative, and Loss inputs to be Positive. Uses of the updated version: Decrease Loss : If positions are in deep loss (e.g. -$5000), and you want the EA to close them at lower loss (e.g. -$2500), you can set the Profit Amount to $-2500. Protect Profit and Loss Trailing : Suppose initially set Profit Amount to $2000, and Loss Amount to -$1000, then profit became $1000. You can then change the Loss Amount to $50 (positive). Suppose after a while, Profit became $1800, then you can...
Hello,

I'm going to test this EA with my Basket Manager, which lacks BE or TS, regretfully, however, since I'm not in front of my charts lately to make adjustments, can you make the Loss Trailing feature automatic, please?


Thank you very much.
  • #1,157
  • Mar 29, 2016 2:15pm Mar 29, 2016 2:15pm
Hello everyone!

One of Abokwaiks EA give me an easy programable idea. It has a very good R/R possibility.

The Rules for short:

10 , 20, 50, 100 ema under each other

RSI (2) goes abowe 90, and when it comes down below (of course wait for candle close, In both cases), we only take the short trade, when the price was not higher then the 20 ema.

Our stop is at the 20 ema. We can use different exit...

It is the basic idea, if somebody could program it, it can be tested. Thanks!
Attached Image (click to enlarge)
Click to Enlarge Name: GBPNZDM1.png Size: 68 KB
  • #1,158
  • Mar 30, 2016 7:28am Mar 30, 2016 7:28am
Another setup.
Attached Image (click to enlarge)
Click to Enlarge Name: EURNZDM1.png Size: 49 KB
  • #1,159
  • Mar 30, 2016 10:25am Mar 30, 2016 10:25am
Hello. Nice system. How often false signals? I see that the indicator increased accuracy and it may cause false signals. Especially at a low timeframe
  • #1,160
  • Mar 31, 2016 4:01pm Mar 31, 2016 4:01pm
Hi Abokwaik,


What do you think about converting this setup to an auto trader? http://www.forexfactory.com/showthre...99#post8804399
0 traders viewing now