2'000 to 20'000 in six months.
Sorry for that title, but im trying to get some attentions :)
Well, first of all, I would like to say thank you to all of you who commented on my last post regarding my previous EA (Previous Post).
It's been a week since then and I coded a new strategy.. It might seem a lame job for you, but for me as a total biginner in programming of any type, im proud of it. I really want to get serious about this, so any other comments and suggestions are again more then wolcome.
This stragegy is coded for EURUSD 1Hour chart and basically analize the strenght and trend of the currency and place an appropriate trade. Its works fine in both ranging and trending market.
I already have my own server to test this on my demo account, but its gonna take couple of months before i got live if at all..
Here is the optimized result for the past six months:
And here is the non-opimized version trading 0.1 lots all the time:
Here is the sumury for the last chart:
And just for really extra effort, here is the last 8 yers test, but honestly, that's not the goal anyway :) Just a confimation that it is trend independent :)
So please if you have any suggestions and anything you think it might be appropriate for it, please comment. I just need some tips or direction of what you think if this can or can't work or if you can see any faults in that..
Thank you a lot for your time and your support!
Jiri
Your modeling quality needs to be fixed. it shows quite a lot mismatch chart errors.
Try re-download data from history center, switch modeling to 'open price' only, run your test, then switch modeling back again to 'every tick' and re-run your test. See if it fixes the errors this time.
This has nothing to do with journal tab.
Thanks Chi! That worked. No more errors and the results are the same :)
If you can try and test the system using M1 data for the most accurate results, you should get Modelling quality of 25% on the strategy tester.
Some ideas to consider.
Your strategy
• It should be based on some simple logical pattern that will continue to be repeated into the future.
• It should be tested on history data with results that are as close to live trading as possible. The prices that would be used in a live trade should be the same as in a historical trade. For example using stop orders will normally guarantee the entry price.
• It should perform consistently on historical testing (at least 5 years), some people don’t consider back testing to useful but if it does not perform in history how can it be expected to perform in the future. Also then you can be confident that the system can handle changing market conditions.
• Consider applying the system to more than 1 currency pair in order to reduce the drawdown risk while maximizing the account growth by applying money management rules that limit the losses. I find this helps smooth out the equity curve rather than trading one pair. I use excel to test different risk scenarios on historical data to understand the possible effect on the account and decide the optimum mix of risk to apply.
• Surprisingly the system doesn’t need to have a high win/lose ratio to make money, even 50% is fine. What is important is the winners must make much more than the losers.
• I think a system can realistically make 10% to 20% per month with an account drawdown between 20% and 30%. As long as it’s consistent. To make a living you need a big enough account balance to generate the income you need. You can’t expect to make $2000 per month with a $2000 balance it would need to be more like $20000 otherwise you would have to take unnecessary risk.
Find a broker
• That is regulated and you can trust.
• That offers unlimited demo accounts that don’t expire so you can test your system properly.
Set up for auto trading
• The system must be set up so that it runs 24 x 7 with minimal interruption
• Look at using a VPS or set up on a managed server, using your home PC could be more risky with power outage and other interference.
Demo test
• Demo test the complete system for as long as possible and compare trades to the strategy tester so you can become confident that it is performing close enough to the historical results.
• Iron out any programming errors you may encounter.
Live test
• Open your live account with a small balance for testing live.
• Compare your live trades with the demo account to ensure things are running smoothly and as expected, you will find live trading results are slightly different to demo.
• Be patient, you have to give yourself the best chance of success and rushing things could hurt your chances.
Thanks derekc1 a lot! It make a lots of sense what you saying.
STRATEGY:
-Its actually based on simple pattern so it should work in any market in any conditions.
-The test is actually done with spread of 3 pips but my broker offers around 2-3 so that could help a bit.
-I believe it performs fine in the past eight years, right?
-Will test it on small mini lots on 2k account for few months i guess, now im testing it on demo account.
I actually got VPS so its connected 24/7, I think that is the best solution for it.
Wish you all the best,
Make sure the data you are testing on is from your broker, you will find that data from the live account may be slightly different to the demo account so the EA should not be too sensitive to the difference.
I suggest you export the optimized (optimize with a fixed lot) test results to excel and play with different risk settings that adjusts the number of lots taken. Then you can see what effect it will have on the drawdown % and you can calculate the average expected % monthly return. At the end of the day the number of pips the system makes is not as important as the % return on investment at an acceptable level of risk (drawdown of account balance). You can make a choice on the level of drawdown you feel you can tolerate. When you know you can expect say a 20% drawdown it will help you through the bad times because you know it's normal.
Utilising excel makes it a bit easier to make the calculations quickly also it gives you the ability of testing what effect adding more currency pairs into the mix will have on the overall system performance, anyway everyone has their own method but this helped me see the bigger picture.
If you want detail about that, please contact me
dian_armando@yahoo.com
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Sorry for that title, but im trying to get some attentions :)
Well, first of all, I would like to say thank you to all of you who commented on my last post regarding my previous EA (Previous Post).
It's been a week since then and I coded a new strategy.. It might seem a lame job for you, but for me as a total biginner in programming of any type, im proud of it. I really want to get serious about this, so any other comments and suggestions are again more then wolcome.
This stragegy is coded for EURUSD 1Hour chart and basically analize the currency momentum and place an appropriate trade. It works fine in both ranging and trending market. Im not traying to make millions, but just a solid steady trading for living :)
I already have my own server to test this on my demo account, but its gonna take couple of months before i got live if at all..
Here is the optimized result for the past six months:
And here is the non-opimized version trading 0.1 lots all the time:
Here is the sumury for the last chart:
And just for really extra effort, here is the last 8 yers test, but honestly, that's not the goal anyway :) Just a confimation that it is trend independent :)
So please if you have any suggestions and anything you think it might be appropriate for it, please comment. I just need some tips or direction of what you think if this can or can't work or if you can see any faults in that..
Thank you a lot for your time and your support!
Jiri