Mithen expert advisor | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

MT4 expert advisor 2 replies

Expert Advisor 8 replies

Vegas Daily MT4 Expert Advisor results 29 replies

Can someone create for me an easy Expert Advisor? 0 replies

Interbank FX Expert Advisor how to program? 15 replies

Attachments: Mithen expert advisor
Exit Attachments

Mithen expert advisor

  • #61
  • Jul 18, 2014 11:09am Jul 18, 2014 11:09am
Quoting Nigor
Hi,Panagiotis! There is a proposal to add another function MinZeroLevelDistantse, function MaxZeroLevelDistanse. It will limit the distance to open an order from level 00. For example: MaxZeroLevelDistanse = 5. This means that if the opening price of the new hour is at a distance of> 5 pips from the 00 level, in this hour on level 00 pending order will not be billed. Thanks in advance!
Thank you for the suggestion. I shall consider adding this option to the next version.
  • #62
  • Jul 18, 2014 12:10pm Jul 18, 2014 12:10pm
hai xux99,

I have tested tested mithen for the whole week.
the EA works great! i run it on a demo account, across EU,GU,EJ,GJ get around 90+ pips nett.
but i didn't run it on full auto, i only enable the EA 30 min before LO (LO is after my lunch hour, I work full-time btw). I close the terminal after a profit or loss and return home and wait for 30 min before NY Open to resume trading.
and I disable the EA 1 hour before news, and turn it back on 30 min after, i didn't use the news filter.

I add the another rule, that is no trade at .0000 level. and the price must be 5pip different from DO for a valid trade.
  • #63
  • Jul 18, 2014 12:52pm Jul 18, 2014 12:52pm
Quoting fxdaytrader_
{quote} ok, set UseCustomLotCalculation=TRUE (default). Then it will use lots = AccountEquity / 1000 ... I think using the equity is better than the balance ... {file}
Thanks
  • #64
  • Jul 18, 2014 12:53pm Jul 18, 2014 12:53pm
Quoting xux99
{quote} You can modify "CalcLots(int x6)" function.
Thanks
  • #65
  • Jul 18, 2014 1:47pm Jul 18, 2014 1:47pm
Hi, I am currently running a backtest from 2013 on EURJPY. It is unbelievably slow on backtest! I started 7 hours ago(!) and it is still 2 months left to backtest!!!
If this is the normal backtest speed I can say that it could never be backtested and fully optimized.

My other EA:s on M1 timeframe I can backtest in about 10 minutes from 2013.

xux99 - Is there anything to do to get faster backtest? It will take months or more on only one pair to do a few tests and optimizing

And the result so far looks like this on a USD10000 account trading fixed 0.10 lots. I started with 1.0 lots but it crashed the account after only 7 months!
http://i62.tinypic.com/wo8pd.jpg
  • #66
  • Jul 18, 2014 4:08pm Jul 18, 2014 4:08pm
Quoting LFN
Hi, I am currently running a backtest from 2013 on EURJPY. It is unbelievably slow on backtest! I started 7 hours ago(!) and it is still 2 months left to backtest!!! If this is the normal backtest speed I can say that it could never be backtested and fully optimized. My other EA:s on M1 timeframe I can backtest in about 10 minutes from 2013. xux99 - Is there anything to do to get faster backtest? It will take months or more on only one pair to do a few tests and optimizing And the result so far looks like this on a USD10000 account trading...
Hi LFN,

The results look very promising.

Keep in mind when testing that we aim for a total of 10 pips collected possible on several pairs during the day. Hereafter we often automatically close the EA because we have achieved our goal for the day. It may be that a pair is less profitable because you only focus on that specific pair and continue trading it while the goal is already achieved.

Just something to keep in mind.

Best regards,

Frietzak
  • #67
  • Jul 18, 2014 4:20pm Jul 18, 2014 4:20pm
Quoting xux99
{quote} Should we take into account MaxOpenTradesAllPairs option and open a trade for not prioritized pairs if total open trades are less that MaxOpenTradesAllPairs-(trading opportunities of priority pairs)? Priority pairs won't completely prevent trading of the rest pairs the will just limit their trades number. It would be nice if other members also make suggestions for priority pairs logic.
Hi Xux99,

Yeah that sounds like an excellent solution. You know that I test this EA with several trading buddies and we all agreed that this is a good solution.

Best regards,

Frietzak
  • #68
  • Jul 18, 2014 4:45pm Jul 18, 2014 4:45pm
Quoting frietzak
{quote} Hi LFN, The results look very promising. Keep in mind when testing that we aim for a total of 10 pips collected possible on several pairs during the day. Hereafter we often automatically close the EA because we have achieved our goal for the day. It may be that a pair is less profitable because you only focus on that specific pair and continue trading it while the goal is already achieved. Just something to keep in mind. Best regards, Frietzak
Hi Frietzak.
Yes I am aware of that. This was just a stress test. Fully sessions trading, not only the first hours of every session. I also disabled the DailyTargetPips and MaxDailyLosses.

At last after almost 10 hours(!) the backtest was finished. I am attaching the backtest and also the setfile. Feel free to upload it to myfxbook for further analysis.
To bad that it takes so long time to perform the backtests. If anyone have some good setfiles, please share them here and I will make it backtested on the pair you want.
---------------------------------------------------------
- Default settings
- Fixed lotsize: 0.10
- Start account $10.000
- 99% Dukascopy tickdata from 2013-01-01
- Real spread
- $7 commission
- Backtest made with TickDataSuite
--------------------------------------------------------
http://i57.tinypic.com/2hzr6mp.gif
Attached File(s)
File Type: zip Mitchen EJ.zip   167 KB | 516 downloads
  • #69
  • Jul 18, 2014 5:28pm Jul 18, 2014 5:28pm
Quoting LFN
Hi, I am currently running a backtest from 2013 on EURJPY. It is unbelievably slow on backtest! I started 7 hours ago(!) and it is still 2 months left to backtest!!! If this is the normal backtest speed I can say that it could never be backtested and fully optimized. My other EA:s on M1 timeframe I can backtest in about 10 minutes from 2013. xux99 - Is there anything to do to get faster backtest? It will take months or more on only one pair to do a few tests and optimizing And the result so far looks like this on a USD10000 account trading...
Try to disable pending orders, it may help. Generally, backtesting scalping strategies does not give very accurate results.
  • #70
  • Jul 18, 2014 5:40pm Jul 18, 2014 5:40pm
imho the solution is: forwardtesting (at least on demo (better: real account on reliable brokers), everything else is wasting of time ...)
so if you have made your first 200-300 USD with the strategy, open a new account with this money and let the ea run (vps), ...

I do not know if the ea already has got this feature: if the strategy says "1 trade per pair and day" just mod. the ea to make only 1 trade per day ...
PM me with coding requests and I'll probably put you on my ignore list
  • #71
  • Jul 19, 2014 2:47am Jul 19, 2014 2:47am
I would not call backtesting useless. It is a good help to check a strategy. I have never put a system live if it not can cope a backtest a few years back.
But this is to interesting to leave.

I will continue to monitor my demoaccount: http://www.myfxbook.com/members/swef...9uBCB2afMMKdhb

One week trading on all 29 pairs. -1088pips.
But it will take a long time before I can see stable results there for choosing right pairs/trading hours/settings and so on. Note that the recommended GJ pair is one of the worst performing pairs this week with -99pips.
  • #72
  • Jul 19, 2014 7:07am Jul 19, 2014 7:07am
Quoting LFN
I would not call backtesting useless. It is a good help to check a strategy. I have never put a system live if it not can cope a backtest a few years back. But this is to interesting to leave. I will continue to monitor my demoaccount: http://www.myfxbook.com/members/swef...9uBCB2afMMKdhb One week trading on all 29 pairs. -1088pips. But it will take a long time before I can see stable results there for choosing right pairs/trading hours/settings and so on. Note that the recommended GJ pair is one...
Hi LFN;

A question about your testing .. Take into account the news red?. Because it is one of the main filters.

Cheers.
  • #73
  • Jul 19, 2014 7:48am Jul 19, 2014 7:48am
Quoting LFN
I would not call backtesting useless. It is a good help to check a strategy. I have never put a system live if it not can cope a backtest a few years back. But this is to interesting to leave. I will continue to monitor my demoaccount: http://www.myfxbook.com/members/swef...9uBCB2afMMKdhb One week trading on all 29 pairs. -1088pips. But it will take a long time before I can see stable results there for choosing right pairs/trading hours/settings and so on. Note that the recommended GJ pair is one...

LFN

thanks for your test

could you please tell us wich is your broker and comission is $7.oo?
  • #74
  • Jul 19, 2014 7:53am Jul 19, 2014 7:53am
Quoting chaveznqoos
{quote} Hi LFN; A question about your testing .. Take into account the news red?. Because it is one of the main filters. Cheers.
No, news is not considered. So this is really worst case results :-)
  • #75
  • Jul 19, 2014 7:54am Jul 19, 2014 7:54am
Quoting Juan007
{quote} LFN thanks for your test could you please tell us wich is your broker and comission is $7.oo?
Broker is Global Prime, commisssion is $7.00.
  • #76
  • Jul 19, 2014 8:23am Jul 19, 2014 8:23am
Quoting LFN
{quote} No, news is not considered. So this is really worst case results :-)

LFN your work its nice

could you considered run the EA with the news
its important about the rules

thanks in advance
  • #77
  • Jul 19, 2014 8:34am Jul 19, 2014 8:34am
Quoting Juan007
{quote} LFN your work its nice could you considered run the EA with the news its important about the rules thanks in advance
Sorry. Maybe I misunderstood you. My demo is considering news as the rules in udines thread and also blocking speeches. It is my backtest that don't consider news filter because I dont think it is even possible in backtest.
  • #78
  • Jul 19, 2014 8:38am Jul 19, 2014 8:38am
Quoting LFN
{quote} Sorry. Maybe I misunderstood you. My demo is considering news as the rules in udines thread and also blocking speeches. It is my backtest that don't consider news filter because I dont think it is even possible in backtest.
ok

muchas gracias
  • #79
  • Jul 19, 2014 8:45am Jul 19, 2014 8:45am
LFN

COULD YOU SPECIFIED THE TRADING TIME PLEASE

its frankfourt open an close on Ny .... ?
  • #80
  • Jul 19, 2014 8:48am Jul 19, 2014 8:48am
Quoting Juan007
LFN COULD YOU SPECIFIED THE TRADING TIME PLEASE its frankfourt open an close on Ny .... ?
Default settings on mitchen ea. I will probably tweak the settings to trade less hours in the future.
0 traders viewing now