Expert Advisors: An Expert Advisor, based on Bollinger Bands

New comment
 

An Expert Advisor, based on Bollinger Bands:

This Expert Advisor is based on Bollinger Bands. It uses trend-following strategy (DEMA) and Bollinger Bands indicator.

It shows a stable profit in Strategy Tester during last 17 years (EURUSD M30).

The Strategy:
  • Open long position: DEМА is growing up and white (bull) candle crossed the Lower Bollinger Band from below to above.
  • Open short position: DEМА is falling down and black (bear) candle crossed the Upper Bollinger Band from above to below.
  • Close long position: Black candle crossed the Upper Bollinger Band from above to below.
  • Close short position: White (bull) candle crossed the Lower Bollinger Band from below to above.

Author: AM2


An Expert Advisor, based on Bollinger Bands

 
hi guys. have anybody heard about forex informer96? i use it and it works perfectly but as far as i know it is based on sun-sensitive market law.. just cant imagine how sun can influence trading.. does anybody have any idea?) cause robot really works but the idea of sun-influence seems to be unbeievable )
 

Good job. It's even amazing that such a simple approach can be profitable. Essentially the strategy should be no better than MACD Sample or something similar, yet it beats such tactics.

Pros:

+ high representativeness of the sample (more than 1000 trades on m30)

+ systematic growth of balance

+ simplicity of the idea

+ works on m30, h1, Daily.

Minuses:

- Works only on EURUSD and only on certain time periods.

- There are no stops. A losing trade can last for weeks.

Ways to solve the problem:

It would be nice to place an effective stop loss, to screen out prolonged unfavourable movements.

The strategy could be tried on less efficient commodity markets (metals, software, energy, etc.).

Often the actual reversal occurs much later than the signal is given. You can try to place an additional indicator indicating the beginning of the actual movement. At the same time, the beginning of this movement can be much higher/lower than the signal point of the bolinger, it should be taken into account.

 

Hi Guys;

Could someone convert this from MQL5 back to MQL4.

My Live account is in MLQ4.

I tested this on an MQL5 Demo acct and it is making money.

This is a very good EA and doesn't need tuning.

Thanks

Trader Girl

 
It's really not bad, but it needs work
 

Hi all, I would like to clarify the author's inputs, sell: high (shift 1) > var1(upper bollingerline ), close < open, close < var1 -----????

If so, then it turns out (on the tester I looked at) it is profitable, but it somehow randomly over-sits, and trades where the price just crosses the khai (shadow) misses.

I have tried many variants with variables = no good stable result. Here it is necessary to add some kind of filter from gap or cunning martin.

If martin, it is necessary to make a coefficient of multiplication of the next lot, the number of orders and exit (if for example bought) on the previous upper fractal (hai) or on the low of the bar on which bought, dolivka not closer than 15 p (should be regulated), please do who knows in coding.

My Skype: rostov-on-don777

 

Hi AM2, thanks for sharing the EA,

I have done some Optimization and for 3 years the same results as you had for 10 :-)

BB


TWF is not used in any calculations it is 3 months left out of any testing , then I run the selected parameters with these 3 month in the end.

See my settings:

BB2

 

Automated-Trading:

An Expert Advisor, based on Bollinger Bands:

Author: AM2

This is a very good EA, is that possible that you add a stop loss?
 

Hi All,

When I run backtest on the same period, I do not have the same results. Can someone explain me why? Here are the result of my backtest. 

Total profit is 26658.15 instead of 33690, but the main difference is on Maximal balance drawdown and Maximal Equity drawdown.

Regards,   

 

 
Files:
bolbands_1999_2010.zip  200 kb
 
and I'm always going negative for some reason.
 
One of the most decent MQL5 EAs I've used in testing so far (and this is big word for me for I generally dislike MQL5, because of american style FIFO...). First of all it's simple, meaning it's not overcomplicated needlessly. Also, it has very clear and well defined entry and exit rules - however it could use a roboust money management. Very WELL done! And thank you!
123
New comment