Indo Run discussion

[suzanna]

I agree, I would LOVE to see a step system based on data instead of a fixed step. It would make this already awesome robot absolutely PERFECT.

 

And thanks for getting this thread going, was missing the discussions on other thread!

:-)

[name000]

I'm testing IR 1.4p (modified v1.4o with added NFP and ADP filters) from 23.11. Managed to make 54% profit, even that EA was not active from 26.11.- 02.12. and 24.12.-11.01. (I forgot to turn it on again :) ). I had no more than 10 open positions (only once, but that was in first few days when set was not completely optimized).

http://www.myfxbook.com/members/name000/indo-r-14p/71855/EMSg0W0ZKGo5rwodGlHs

http://www.myfxbook.com/widgets/71855/large.jpg

 

Here you have set and modified v1.4o:

http://www.mediafire.com/?z7lo37kdm7qrom7

But, even thought this is quite 'safe', I don't think that IR is still safe enough because no matter which setup you have, there is always a chance to blow your account as it is almost happened to haynese1975. I would rather invite you to read this thread:

http://indo-investasi.com/showthread.php/11454-IndoRun-exit-solution

I'm not 100% sure that this can be right solution, but at least we can try and exchange opinions and maybe we can make something good out of that.

[JasonKwok]

like any other grid trading EA, it's always better to withdraw or transfer the profit out every week.

 

I'm running Indo-Run on Alpari UK Micro account and every weekend, I will transfer the week's profit to another Alpari UK account that is not trading with any EA. The transfer can be done online easily with Alpari UK.

 

My objective of moving the profit out every week is to recover the initial capital before any disaster that will wipe out the account. After the capital has been recovered, then for subsequent profit, I will withdraw halve and leave the other halve to growth the account.

 

We do not know how the EA will behave and best is to recover the capital and subsequently, it will be risk free trades.....;-)

 

Please do not forget to contribute some profit to the developer. ;-)

[suzanna]

Back test 1st June till 25th October 2010. All the DD dips on this chart are caused by news events. Even though news filter is turned on the back-test data doesn't recognise the news filter thus they can be avoided.

http://img825.imageshack.us/img825/2048/1622010to25102010.gif

 

Which settings are these please? Pair? Time Frame?

[fabfive]

Hey Tomislav,

 

have you been using this setfile live? 24h? since when?

 

The backtest quality shows 60%.

Edited February 6, 2011 by fabfive

[suzanna]

Second part of the back-test 25-10-2010 to 6-2-2011. All the draw down dips are traceable to news events so can be avoided according to FXCM data (except one on New Years Eve 5th dip from the end which isn't that bad really). News filter doesn't operate while back-testing. Avoids May 6th mini-crash. Makes good profit and a very good chance it won't crash your account.

 

Set File GBPUSD 15 min TF

http://www.4shared.com/file/SbcsOX0s/indo_15_dynamic_profit_momentu.html

 

http://img823.imageshack.us/img823/2943/2512010to622011.gif

 

So instead of additional filters to expats settings, you are using only the ma and momentum? Interesting. Did you do backtest comparisons with expats set file and using ALL filters (his and yours)?

 

Thanks for sharing this with us!

:-)

[suzanna]

Number 15! Now I can PM.

:-)

[suzanna]

Tomislav, I think this is the first robot/settings (YOUR settings) that I have ever been able to backtest right thru 2008 without a crash. GREAT job!

[fabfive]

Have you backtested my settings suzanna?

[mclane]

Tomislav, I think this is the first robot/settings (YOUR settings) that I have ever been able to backtest right thru 2008 without a crash. GREAT job!

 

suzanna can you post your backtest result. thx

[suzanna]

Have you backtested my settings suzanna?

 

Not yet. It takes my computer FOREVER to run it and I was just really interested in the dynamic profit on Tomislav's. I will run yours today.

[suzanna]

http://interestedpeople.com/tomislav%202008.gif

 

Details: http://interestedpeople.com/tomislav%202008.htm

 

I did change a few things, like no trading Friday and end of month, and I ran it from Mid Jan to Mid December, which is what I would have done live. But all other settings are Tomislav's set file.

Edited February 9, 2011 by suzanna

[fabfive]

Ok thank you suzanna.

I wish I could backtest it myself but strangely all my MT4s dont do it properly.

[suzanna]

I have months of missing data in 2010... I think everyone does. But you can try from about September to present ok. I am testing 2008 right now and so far so good, but I have a feeling it will crash in August 08 as most robots do. Will let you know.

[fabfive]

Ive got full 2010 tick data from dukascopy here... but my metatrader just doesnt backtest

[suzanna]

Actually, I ran the test on 2010 to present, not 2008. It has been running all day and still not done yet. BUT it is looking very good, despite the holes in the data. I will run 2008 as well and post both results.

 

I will also test a few other pairs and see if it is as good with those.

[fabfive]

Great suzanna. Afterwards we can decide which set file to choose.

[didier]

Great suzanna. Afterwards we can decide which set file to choose.

 

Hi fabfive,

 

Thanks again for your mod (4order indorun) ;)

I noticed that there are differrences between your setfile and your settings

posted here at the first page:

 

your settings:

- TradeMondayAfterNFP false

- TradeADP false

 

your set file:

- TradeMondayAfterNFP true

- TradeADP true

 

Thanks for your attention 8-)

[fabfive]

I will post an updated setfile after the backtest.

[suzanna]

I am just going to bed. The test fro 2010 crashed in Nov. And for 2008 it crashed in Feb, then August and I stopped there. Will upload and post images/results tomorrow for you. So far Tomislav's set file produced better results with NO crashes. That said, except for the crashes, Fab5's set traded really well, so by using it live and avoiding news events, it should be very much ok too and total result about the same. Edited February 9, 2011 by suzanna

[fabfive]

Hey

My backtesting works now. Seems like a miracle..

annnnnnnnnnd I got Dukascopy 99% modelling quality backtest data for 1.3.2007 - 31.12.2010.

I will post the results.

[massa6000]

Fabfive i got this great tick data here too... and i wanna tell you something...

I tried made optimization here with 1.3.2007 - 31.12.2010 data and dont worked... I asked Birt what's the problem, and another user told me to use 2008 - 2010 and it worked...

So i dont know if the problem its the data more than 2gb or if 2007 data have problems... but that's just for optimization... backtests without optimization runs perfectly...

Another trick is use brokers that have 5 digits... I've tried a broker with 4 digits and i had problems.

I'm trying to optimize Indorun too

Let's do it together ;)

[agent6]

hi i also discovered that problem. it is not possible to use tickdata form 2007-2010 in one file, it gets to big you might be able to use it but when i tried it worked from 2007 to 2008, but not with other years, is there anyone who knows how to get around this 2gb problem? merging maybe?? please help by the way , i bought a guide that explains in, really good for me as i am new to forex, you find it on birts page! good luck !

[suzanna]

Sorry guys. I haven't had time to upload the reports. But basically it is all good, except for the crashes. The fabfive settings will crash from time to time. Tomislav's will too, so he says, even though it never did in my testing. So you just need to keep your eyes on the news filter and if there is a real world event making the markets jumpy, I'd go in and turn the robot off until things settle down. Remember, robots are a trading tool, they aren't trading gods!

[verde]

The indo 1.5 dynamic profit set that tomislav posted doesn´t open any trade, anyone wit the same issue? I´m using alpari demo