Try My EA! Need Group Feedback... | Page 31 | Crypto Craft

Options

Go to
Permalinks
Print Thread

Related Instruments

Gold/USD

Similar Threads

USD/TRY and EUR/TRY 12 replies

Fibo Group Feedback 23 replies

Need feedback on my trading breakouts system 8 replies

One World Capital... Need feedback 2 replies

Need feedback on FXCM Educational workshops 3 replies

Attachments: Try My EA! Need Group Feedback...
Exit Attachments

Try My EA! Need Group Feedback...

  • #601
  • Dec 22, 2014 3:39pm Dec 22, 2014 3:39pm
Quoting Thepipster
{quote} gas is $1.96 a gallon here...
Surely, you mean petrol.
It's around £1.20 per litre here, which makes it about four times more expensive (!?).
Should we start an import/export business then?
mind over matter
  • #602
  • Dec 22, 2014 3:55pm Dec 22, 2014 3:55pm
http://www.kitco.com/images/live/nyg...07787682455684

Quoting kosmo
{quote} Surely, you mean petrol. It's around £1.20 per litre here, which makes it about four times more expensive (!?). Should we start an import/export business then?
I guess...we call it gas down here. We're just good ol' country folk.
https://encrypted-tbn1.gstatic.com/i...gncwFyqvhTUK5w

The cultural\colloquial spectrum on this site is fascinating to me, sometimes.
Too close for pips, switching to points.
  • #603
  • Dec 22, 2014 4:10pm Dec 22, 2014 4:10pm
I stoped testing for now may be later..
  • #604
  • Dec 23, 2014 12:30am Dec 23, 2014 12:30am
@ThePipster,
would you mind to explain in details technically for each and every parameters in the screenshot below?

Why I'm asking this is because of some broker, doesn't reflect the same as in Armada. I know this could be because of different spreads, but if you can explain it, I can try to find another settings that fits with another broker than Armada.
If you see in the 1st screenshot below, In Armada (left), the order created was so tight, but different in another broker (right, still loose after changing some parameters).
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot2.png Size: 53 KB
Attached Image
If you ask me to code/fix your EA... it's probably not for free...
  • #605
  • Dec 23, 2014 7:56am Dec 23, 2014 7:56am
Quoting radityo.ardi
@ThePipster, would you mind to explain in details technically for each and every parameters in the screenshot below? Why I'm asking this is because of some broker, doesn't reflect the same as in Armada. I know this could be because of different spreads, but if you can explain it...

It is because of the spreads. The buffer and gridstep adds more distance between orders. If you use value 10, it adds $0.10 to the distance.
Too close for pips, switching to points.
  • #606
  • Dec 23, 2014 9:57pm Dec 23, 2014 9:57pm
Quote
@Pipster: Just wondering if it would be possible to stop out the trades instead of hedging them. The reason would be to save the double hit on spreads that a hedge imposes once the hedge is closed, as compared to simply closing the trade and not hedging.

Here's a modified version of Pipster's EA to address this issue. I'm using version 765.

When a trade is opened from the pending grid pool it's opposite number is deleted. No hedging should occur. A stop loss is in the place of deleted side for the opened trade, these are put in on initiation of the grid. This version should function the same as the original, but should save half the commission and spread cost.

Of course, with no hedge, it won't be possible to step out of the hedge grid one side at a time, such as Pipster just did.

A comment is put in the place of maxloss to say what gain is being sought.

The default grid spacing, etc. are in line with Kosmo's settings.

A flag is placed in the inputs to allow the time check to be bypassed. ("UseTimeOfDay") Set this to true to make the EA wait on the clock.

Any errors or loss of functionality is purely my fault. I have it working and it seems to be doing fine. I'll put up a trade explorer for this version too.
Attached File(s)
File Type: ex4 Steady Gold NoHedge.ex4   20 KB | 342 downloads
File Type: mq4 Steady Gold NoHedge.mq4   12 KB | 397 downloads
A good man knows his limitations ~ Clint Eastwood
  • #607
  • Dec 23, 2014 10:08pm Dec 23, 2014 10:08pm
Quoting Thepipster
{quote} It is because of the spreads. The buffer and gridstep adds more distance between orders. If you use value 10, it adds $0.10 to the distance.
Thanks, though.

Is the pic below the best illustration for your EA?
Attached Image
If you ask me to code/fix your EA... it's probably not for free...
  • #608
  • Dec 24, 2014 8:14am Dec 24, 2014 8:14am
Here's a chart using the non-hedging version. Just going sideways in a narrow range, hollowing out the grid. No open positions at the moment. Since the sell positions are put on after the buy positions in the code, they overwrite the buy text on the screen. The buys are there though.

Attached Image (click to enlarge)
Click to Enlarge Name: XAUUSDM15.png Size: 89 KB
A good man knows his limitations ~ Clint Eastwood
  • #609
  • Dec 26, 2014 1:08am Dec 26, 2014 1:08am
I put up the code for a version that doesn't hedge the trades, but I put in a request to have the post removed. I had used a different method to store the starting balance on the account and it caused a reset when MT4 was turned off and back on. Not good.

I have put Pipster's method back in but I'll test it a while before posting the code. Sorry for the confusion if anyone is trying it out. The difference is in using the Get and SetGlobalVariable functions in MT4, which are stored for a while on the computer.

Got hammered a little on the open - gapped open past my stops on the two open sell orders on the non-hedged version. Guess that's a weakness in having no hedge.
A good man knows his limitations ~ Clint Eastwood
  • #610
  • Dec 26, 2014 7:55am Dec 26, 2014 7:55am
Quoting radityo.ardi
{quote} Thanks, though. Is the pic below the best illustration for your EA?
No, that's a standard hedge you're looking at. This EA only has stop entry orders.

As everybody can see, we had a big gap up after the holidays, so my closing all the buy trades caused a margin call, but the EA will still be trading from a smaller balance. Of course, if this had been a live account, I never would have closed those trades, but just added more on each side, and sat and waited. I was inpatient, and wanted the EA to be able to continue cycling so people could see how it works. I will leave it running at the lower balance so that everyone can see the longer-term results, until someone can implement the TP\SL idea.

Thanks, dgodfrey, for your efforts, on coding SLs so we don't have to close so many open orders. That should help protect profits.

There are still trade explorers on the first page that show the EA doing very well.
Too close for pips, switching to points.
  • #611
  • Dec 26, 2014 8:06am Dec 26, 2014 8:06am
My one currently kickin' a$$
  • #612
  • Dec 26, 2014 9:49am Dec 26, 2014 9:49am
Quote
I never would have closed those trades, but just added more on each side, and sat and waited.

I guess that's just done manually? I dumped my trades when I got boxed, but I should have just maybe added to it as you say. Just closed out my trades at a small profit and opened it back up on the standard hedge - I had the wrong lot size on it, too small.

The non-hedge is working too, closed out for just over 1% profit as requested and reopened.

That gap was good to you, Dambuster!
A good man knows his limitations ~ Clint Eastwood
  • #613
  • Dec 26, 2014 10:29am Dec 26, 2014 10:29am
Quoting dgodfrey
{quote} I guess that's just done manually? I dumped my trades when I got boxed, but I should have just maybe added to it as you say. Just closed out my trades at a small profit and opened it back up on the standard hedge - I had the wrong lot size on it, too small. The non-hedge is working too, closed out for just over 1% profit as requested and reopened. That gap was good to you, Dambuster!
So, is your version doing ok, then, with keeping the original equity target, despite closing trades to SL?
Too close for pips, switching to points.
  • #614
  • Edited 9:36pm Dec 26, 2014 10:41am | Edited 9:36pm
Yeah, it was working before that way, but I had to put back in the global variable functionality to maintain an accurate starting balance when the market closed and I shut down the computer. After I put back the global variable I deliberately closed and opened it a few times, turned off the MT4, etc to check it. It's working like your original program now, as far as holding onto the starting balance. It was working before without these latest changes as regards to taking SL's.

I have it running on an explorer called "Steady Gold No Hedge" separate from another one with your original code.
A good man knows his limitations ~ Clint Eastwood
  • #615
  • Dec 27, 2014 9:17am Dec 27, 2014 9:17am
@Pipster and/or dgodfrey

Seeing how Wednesday (24.12) and Friday (26.12) played out, I have a thought/request/suggestion.

For the original version, would you be able to code in a visual aid, in the form of two horizontal lines (custom colour), showing the BE area and the target area on the chart, for every cycle? For those who, like myself, will be using the EA as a semi-auto tool, it will be an invaluable help in decision making, based on PA.

Have a look at the picture. I run this test without manual intervention, but if I was using it for live trading, the third bounce off the ascending support line (marked as 'BE area') would be my decision point to reload the cycle. Why? Because I could see that PA was very reluctant to go lower; it formed a 123 pattern around the support line and broke out to the upside. Now, seeing all this action so close to my BE point, with no chance to reach the target, it would be a perfect place to reload the cycle and grab the rally instead, with minimal DD.

I hope you understand and are able to add the feature. Many thanks in advance.
Attached Image (click to enlarge)
Click to Enlarge Name: betarget.png Size: 60 KB
mind over matter
  • #616
  • Edited Dec 28, 2014 11:59am Dec 27, 2014 3:04pm | Edited Dec 28, 2014 11:59am
I had been thinking about those numbers, Kosmo, just in the vein of reasonable expectation to get a return. That's why I closed out that box, there wasn't any reasonable expectation for gold to get back over $1400 anytime soon. Do you have any idea about how they should be displayed, though? I was thinking that if they were just lines and close by they would probably get masked by the grid itself. If they were far away (like my 1400+) you wouldn't know where they were. And there will be two sets of BE's and targets, of course, on the buy side and the sell side, unless all the orders on one side are already filled. Then there would only be one way to make a profit - on the filled side.

They would also necessarily be hypothetical, since they would assume perfect fills with no slippage in the trades that might happen on the way to those levels. Kind of tricky, in that getting to the profit line, for instance, might involve hitting additional stop orders, which would alter the profit line. What I'm trying to say is that although the breakeven or profit target might exist at a particular price point now, by the time you got there it might have moved to another price point in some scenarios. With respect to your example chart that's probably not the case for the sell side, but probably would be for the buy side.

I'll see if I can tinker with it anyway. I'll just stick in the custom color lines and we'll see how that works. Sounds like fun
A good man knows his limitations ~ Clint Eastwood
  • #617
  • Dec 27, 2014 5:54pm Dec 27, 2014 5:54pm
Quoting dgodfrey
I had been thinking about those numbers, Kosmo, just in the vein of reasonable expectation to get a return. That's why I closed out that box, there wasn't any reasonable expectation to wait on gold to get back over $1400. Do you have any idea about how they should be displayed, though? I was thinking that if they were just lines and close by they would probably get masked by the grid itself. If they were far away (like my 1400+) you wouldn't know where they were. And there will be two sets of BE's and targets, of course, on the buy side and the...

I'm glad we share the idea, dgodfrey.

What I have in mind is just simple horizontal lines. Even if the lines themselves get masked by the grid, still the price level will be highlighted on the price scale, and that's all I need to see.

You actually made me realise there would need to be two target lines, or target projections. But I think there will only be one BE line - once trades get open, no matter what the proportion between the number of buys and sells, there will only be one common BE level for all of them. And of course, it will be moving up and down (as will the target projections) as price fluctuates and subsequent pendings get hit, so the lines need to auto-update to show the current BE/target levels.

I imagine that the lines will reflect, and be based on, whatever chosen settings. I don't mind if they only show approximate levels (the more accurate the better, of course); it's just meant to be visual aid. Based on current PA as well as the number of trades currently open on each side, I simply want to have a quick reference point to better picture the probabilities.

The chart I attached in my previous post was very rudimentary, indeed. It didn't really show the whole story, i.e. by the time price reached the marked BE area, I had more sells than buys open, and the actual target was just a few levels below the orange support line. It was clear, though, that price wouldn't go any lower. As it was also clear that it started changing direction, had I had the BE/target lines in place, it would've been much easier to see it was high time to reload.

I'm sure you'll come up with something clever.
mind over matter
  • #618
  • Dec 27, 2014 11:32pm Dec 27, 2014 11:32pm
Might the attached be similar to what you are looking for?
Attached File(s)
File Type: mq4 1 Average Price.mq4   4 KB | 375 downloads
  • #619
  • Dec 28, 2014 11:57am Dec 28, 2014 11:57am
Quote
Might the attached be similar to what you are looking for?

Looks like it might be a step that way. I've got the coding done, I think. Just need to test it to see if it's really going to work or not. Only puts up the numbers in the comment block at present, I'm not really sure you can do indicator stuff from within an EA like put objects on the screen, I'll have to check on that.
A good man knows his limitations ~ Clint Eastwood
  • #620
  • Dec 28, 2014 6:26pm Dec 28, 2014 6:26pm
Quoting dambuster
Might the attached be similar to what you are looking for? {file}
Very useful, thanks!

All we need now is the target projections.

@dgodfrey - maybe it would be easier to modify the indi to show targets, rather than try and implement it in the EA... (?)
mind over matter
1 trader viewing now