Simbasystem-gbpusd - page 86
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Simba thanks
can you give the settings to gbpjpy ?
Simba thanks can you give the settings to gbpjpy ?
Identical with gbpusd.
fxpertno1
I have been following this thread. I am not familiar with MQL at all.I wish I could be of more help to you all. Here is some data already on MT4.
This fellow has posted M1 data back to 2001 for the 4 majors here:
http://www.michaelsaganski.com/downloads/
You can fill in the gap to 2007 by downloading the M1 data provided by Alpari here:
http://www.alpari-idc.com/en/dc/databank.php
Hope this helps and speeds up the process??
__________________Hi,
Thanks,seems it is solved but when talking about data you never know,so,thanks for posting the links
Regards
Simba
canados
Hi canados,
JMA is not an indicator we actually use,though it has potential,I would attach it but I have deleted it from my pc,so,if you do a search in this Forum ,with search function,above to the right,just below where you see "Welcome canados"..;)... you will easily find it.
JMA13 stands for JMA,13 periods...
Regards
Simba
che
Simba thanks can you give the settings to gbpjpy ?
Hi che,
As Frantacech explains in his previous posts,the settings are the same for both pairs,settings were derived from in sample tests of gbpusd and tested out of sample on gbpjpy with good results(not so for usdchf,usdjpy,gbpchf),though probably there is room for improvement here we feel confident enough to start forward testing on both pairs.
You can go to Frantacech `s post 804 in this thread where he has posted the links to 2 concurrent tests we are running,one"official" are the official settings ,same for both pairs,the second"b" test is the"official+break even and small profit stop" where there is an additionaltrailing stop strategy that is explained in previous Frantacech`s posts,that we feel can have potential..
Regards
Simba
Next Steps
Hi ,
Ok we have arrived at a junction point and there are several different roads we may want to take,IMHO we have the following choices..while keeping with the actual forward test,of course..
1-TEST OFFICIAL+B VARIATION OUT OF SAMPLE..WITH THE SAME SETTINGS(obviously)..SO TEST ON JAN01 TO JUN 04 FOR CABLE AND GBPJPY
2-IMPROVE THE SYSTEM FOR ALTERNATIVE SETTINGS FOR USDCHF,USDJPY,EURJPY,EURUSD
3-IMPROVE THE SYSTEM ADDING THE MISSING PART,THE PIVOTS,AND WITHOUT CHANGING THE OFFICIAL SETTINGS ,ADD SOME FILTER LIKE "if price x pips below the pivot-for a sell-launch a limiyt order at pivot,instead of actual market order.."
4-START WORKING ON MM AS A MULTIPLIER..I WOULDN`T LIKE TO DO SO,UNLESS WE HAVE COMPLETED PART1 AND WE HAVE MORE OUT OF SAMPLE GOOD RESULTS
I would like to know your opinion regarding which is your favored road,the reasons why,so that we can discuss(amicably..;)) and decide
Regards
Simba
Hi guys,
I have solved our data issue. So let me sum up what fxpertno1, PeaceLover, and Simba figured and what I have figured to have a complete manual as to how to have M1 data from January 2001 till today, therefore being able to test EAs with 90% modelling quality over this period of time
Step 1:
Download data from http://www.guiglionda.com/forexdata.aspx. You have to register but you'll be able to download data for all majors from January 2001 till like fall of 2006 (applies to GBPUSD, I haven't tried the others). Also please use Internet Explorer because the site is not optimized for other browsers and I had trouble finding the link with Mozilla Firefox. All is in TXT.
Step 2:
Download the latest data (from June 2004 till today) the way you probably already know from http://www.alpari.co.uk/en/dc/databank.html. Those are in HST.
Step 3:
Delete all HST files from /history folder. Then open your MT4 working offline, press F2, and import the TXT file you downloaded from guiglionda.com. Make sure you selected to import TXT files. Close your M4 and reopen. When you go Open Offline, you should see the data over the period I mentioned above. This is just a simple import, all details have been discussed in other threads.
Step 4:
In History Center (F2), hit Export, and export the data as CSV. Name the file e.g. GBPUSD_part1.csv. This is the old guiglionda.com data converted to CSV.
Step 5:
Repeat steps 3 and 4, only this time using the HST file downloaded from Alpari. Save the file as GBPUSD_part2.csv, this is the latest Alpari data converted to CSV.
Step 6:
Back up those two files in case you spoil something. Open GBPUSD_part1.csv in the classical Windows' Notepad application, or any other text editor you are familiar with. Do not use Excel (it will try to force you to do that). Give your computer time -- not only now but all the way through. It will be dealing with 1M to 2M records in its RAM, not exactly a fast thing... When the computer e.g. doesn't change your cursor after pointing the mouse at different icons or texts, it means it's busy and you should not bother it.
Step 7:
Scroll to the line with data June 16, 2004 9:11. Delete ALL lines BELOW it INCLUDING this line (can take some CPU time).
Step 8:
Leaving this open and unsaved, open GBPUSD_part2.csv in text editor (CPU time). Press Ctrl+A to select everything, then Ctrl+C to copy to clipboard (CPU time). Switch back to GBPUSD_part1.csv, go to the last line, hit Enter to create a new one, and hit Ctrl+V to paste the clipboard (CPU time). Close GBPUSD_part2.csv.
Step 9:
Double check GBPUSD_part1.csv, get rid of any possible empty lines etc., just make it look nice and clean... Also check the lines around June 16, 2004 9:11 and make sure there's no duplicite lines.
Step 10:
Save this file (CPU time) and close it. BACK IT UP AGAIN! This will be your main data file. Basically you can delete the others. Make your computer settle down after such operations.
Step 11:
Repeat steps 3 and 4 again to import GBPUSD_part1.csv. Close MT4, reopen and you got your data
When you download new Alpari data next time, I think that it would be best to convert them to CSV, get rid of all but the latest lines (those you don't have in the main file mentioned in step 10), and append these latest lines to the main file. Simply keep updating the main file; it will be best CPU-timewise.
If you have some software for appending files, you can use that.
I hope we will have more testers now
It will be good for us and for you too. If you trade without even taking the least effort to do your backtests... please do so, you'll be Simba's food... and you'll be my food too when I'm an experienced trader. The more of such pricks, the better odds for me eat them...
Enjoy
Hi ,
Ok we have arrived at a junction point and there are several different roads we may want to take,IMHO we have the following choices..while keeping with the actual forward test,of course..
1-TEST OFFICIAL+B VARIATION OUT OF SAMPLE..WITH THE SAME SETTINGS(obviously)..SO TEST ON JAN01 TO JUN 04 FOR CABLE AND GBPJPY
2-IMPROVE THE SYSTEM FOR ALTERNATIVE SETTINGS FOR USDCHF,USDJPY,EURJPY,EURUSD
3-IMPROVE THE SYSTEM ADDING THE MISSING PART,THE PIVOTS,AND WITHOUT CHANGING THE OFFICIAL SETTINGS ,ADD SOME FILTER LIKE "if price x pips below the pivot-for a sell-launch a limiyt order at pivot,instead of actual market order.."
4-START WORKING ON MM AS A MULTIPLIER..I WOULDN`T LIKE TO DO SO,UNLESS WE HAVE COMPLETED PART1 AND WE HAVE MORE OUT OF SAMPLE GOOD RESULTS
I would like to know your opinion regarding which is your favored road,the reasons why,so that we can discuss(amicably..;)) and decide
Regards
SimbaI like this timetable... I'll try to do the EA as soon as possible. I've already started but didn't have time yesterday or today... pray it will be tomorrow
It would make my job a lot easier if we could just start with following:
When there's a buy signal, place a limit order. The limit would be the nearest supporting line below current price (pivot or one of the three supports).
When there's a sell signal, place a limit order. The limit would be the nearest resisting line above current price (pivot or one of the three resistances).
Left to be decided:
How long will the limit order be valid?
What if the price is slightly below, let's say, pivot (long)? Should we place a limit at the line below, which is now very far, or should we take the trade right away? What would be the maximum distance from the line above to take it right away instead of placing limit? (All for long, opposite for short.)
Please comment, everyone
Malcik Pivots
I like this timetable... I'll try to do the EA as soon as possible. I've already started but didn't have time yesterday or today... pray it will be tomorrow
When there's a buy signal, place a limit order. The limit would be the nearest supporting line below current price (pivot or one of the three supports).OK
When there's a sell signal, place a limit order. The limit would be the nearest resisting line above current price (pivot or one of the three resistances).OK
Left to be decided:
How long will the limit order be valid?EXTERNAL VARIABLE "TIME LENGTH" or whatever you want to name it for testingWhat if the price is slightly below, let's say, pivot (long)? Should we place a limit at the line below, which is now very far, or should we take the trade right away? OPEN OPTION to test for best....would be the maximum distance from the line above to take it right away instead of placing limit?OPEN OPTION same as above(All for long, opposite for short.)
Please comment, everyoneHi Malcik,
My comments are above in blue..
Regarding your previous post,I hope you don`t get an indigestion..leave some food for other traders too
..i
If people don`t want to test ,leave them at it ,their probable primary motivation for trading is not making money but recover their lost belief in fairy tales..Leaving goblins aside(bluto,no pun intended believe me
) since I respect bluto`s work,which is exceptional..we have...
..Elves,trolls,witches,ogres,giants and demons(wow...how many names for new EAs..I can foresee them in ebay in a few days just with the same reality base as the fairy tales they are based on)...you know,a great deal of so-called fairy tales do not feature fairies at all..and a great deal of so-called profitable EAs/Systems do not feature any profit at all nor any systematic out of sample testing that can,at least improve the odds of saving Little Red Riding Hood from the Wolf...
These people just don`t yet realize that market reality/their pet indicator is not their fairy godmother but more like THE wicked stepmother,and that the only way to be able to differentiate between the 2 of them is by sytematic testing..
Don`t wake the sleeping beauty(wow a new name for an EA..tough I prefer SnowWhite at least there is some reality contained in its definition ),let her dream..;)
Regards
Simba
Hi canados,
JMA is not an indicator we actually use,though it has potential,I would attach it but I have deleted it from my pc,so,if you do a search in this Forum ,with search function,above to the right,just below where you see "Welcome canados"..;)... you will easily find it.
JMA13 stands for JMA,13 periods...
Regards
Simbathanks a lot