Is it possible to write a profitable EA based on Technical Indicators?

 

Hey everyone,

is it even possible to wirte an EA based on technical indicators, that generates profit over a long period? i don't mean that every trade is a success but that it generates a profit over several years. Looking at the Championchip 2012 i realize that only 20% of the participants generated profit. and even if I look at the current #1, JPAlonso. He made a huge profit in october. However if the championchip would have started on 24. october, he would have only generated loss since then! Is it possible, that it is pure luck, that some participants have got such good results? I miss curves, that earn (little) profit, but constantly over a longer period! 

I mean, if you play roulette, you'll have some winner and some loser at the table, maybe 40% winner and 60% loser after a few rounds. the winner only won because of luck and if everyone plays long enough, everyone will lose their money exept the bank. isn't trading with EA similar?

i'm asking myself the question, because I want to write my master-thesis about this topic and I'm afraid to 'waste' my time on something, that cannot work. What do you think about?  

 

everything is possible.

 

I do not think this is a waste of time, automated trading systems are a relatively recent area of study (i say it is relative because every year we discover how little we know and how much can be done). As onlysolo said, "everything is possible", that means you (or someone else) can discover the holy grail (wich is the main reason most people are here) or simply all the strategies you develop fail. How can that help to your master-thesis? In a lot of ways (2 actually that i can think of hehe):

  1. Develop a succesful strategy and prove in your thesis that it is possible and you could even talk about how much is there to be done or to be discovered/invented or optimized. Talk about indicators used, why you selected those parameters and what values those parameters make your strategy to fail and why. If you're stuck with a long term investment period you'll have to base your studies in backtesting though.
  2. If your strategy fails you can make an interesting case of study and base your thesis on why automated trading systems fails, how to fix and optimize them, how should a perfect market be for a specific strategy, why it works in backtesting but not on realtime account. Then make references to types of strategies and how much they last before breakdown and why. Expose performance from manual trading and automated trading and find correlations.

Don't know if you need to base your thesis on a specific EA or you can make a more generic approach and talk about traders (their ages, aspirations, techniques, etc) and their relations/experiences with their trading systems, then classify those traders and their strategies...finally you could possibly make a strategy ranking based on performance over a period of time.

There's tons of new information to be gathered that could keep you writing for years. That's what i think, thanks for reading.

 
A trading championship has not much to do with real trading.  In a championship you try to be as aggressive as possible and you depend on luck (a series of winning trades) very much.  This is mainly because you know that everyone is going to trade aggressive and if you dont, you can not win.  In a normal environment (trading) for long term succes you will use lower risk and not overoptimize your ea to recent data.
 
Weedjo:

Hey everyone,

is it even possible to wirte an EA based on technical indicators, that generates profit over a long period?

... 

Yes, it's possible since a long time ago. please read this first, maybe I'll write later : 

1. https://en.wikipedia.org/wiki/Quantitative_analyst 

2. https://en.wikipedia.org/wiki/Algorithmic_trading 

 

Quantitative analyst - Wikipedia, the free encyclopedia
Quantitative analyst - Wikipedia, the free encyclopedia
  • en.wikipedia.org
A quantitative analyst, or quant is a person who specializes in the application of mathematical methods, such as numerical or quantitative techniques to financial and risk management problems. Similar work of industrial mathematics is done in most other modern industries, but the work is not always called quantitative analysis.1 See List of...
 
see poll in this forum: https://www.mql5.com/en/forum/8357
 
Weedjo:

Hey everyone,

is it even possible to wirte an EA based on technical indicators, that generates profit over a long period? i don't mean that every trade is a success but that it generates a profit over several years. Looking at the Championchip 2012 i realize that only 20% of the participants generated profit. and even if I look at the current #1, JPAlonso. He made a huge profit in october. However if the championchip would have started on 24. october, he would have only generated loss since then! Is it possible, that it is pure luck, that some participants have got such good results? I miss curves, that earn (little) profit, but constantly over a longer period! 

I mean, if you play roulette, you'll have some winner and some loser at the table, maybe 40% winner and 60% loser after a few rounds. the winner only won because of luck and if everyone plays long enough, everyone will lose their money exept the bank. isn't trading with EA similar?

i'm asking myself the question, because I want to write my master-thesis about this topic and I'm afraid to 'waste' my time on something, that cannot work. What do you think about?  

I don't think trading is equal or even similar with gambling - not even close ;D. Those rich list from trading the market (Warren Buffett, George Soros, et al) are never gamble the market.

Trading, mainly is buy and sell something with a plan, while gambling does not buy or sell something even with a plan.

Say for example a fruit seller who sells Apples, Oranges, Bananas, Grapes, etc.. Fruit seller maybe buys fruits on Friday, hoping he can sell it on weekend before the fruits goes rotten. On weekend his Apples, Oranges, and Bananas are sold out, but not the Grapes. So on Monday he have to sell the Grapes below its buying  prices - unless he wants a rotten Grapes - so he sells the Grapes at a loss. At least the Fruit Seller can make the capital back and keep his fruit business running.

If JPAlonso is a Ice Cream seller. I don't think he will make much of profit (will certainly makes more loss) selling ice cream in Winter compare than selling ice cream in Summer.

In gambling, a gambler buy and sell nothing. Well, actually a gambler create himself a hope for himself and sell his money to buy that hopes.   

In Fruit and Ice Cream example, a marketing and selling is a good plan before trading. In trading the market and exchanges, good reading on indicators and all trading possibilities is one-third good plan, making firm trading decision is another one-third good plan and another third lies in executing it.

A good EA is the one which reads the indicators and all possibilities, "read" it's plan and executing the plan.

Documentation on MQL5: Standard Constants, Enumerations and Structures / Environment State / Symbol Properties
Documentation on MQL5: Standard Constants, Enumerations and Structures / Environment State / Symbol Properties
  • www.mql5.com
Standard Constants, Enumerations and Structures / Environment State / Symbol Properties - Documentation on MQL5
 

thank you all for your great response! I had my doubts, that's why I asked.

I think I'll make it like TripleHeinz said: Even if my studies would fail, there would be additional value to science in this field.

 
I think its very possible! but of course that EA should understand about fundamental and not only Technical only, cause i believe that fundamental is the key of success of profitbale EA, but to be honest if you think EA scalper can make profitable for long term I doubt about that, maye for 3 month it can be really profitable, but not for long :) just my 2 cents
Why Is MQL5 Market the Best Place for Selling Trading Strategies and Technical Indicators
Why Is MQL5 Market the Best Place for Selling Trading Strategies and Technical Indicators
  • 2012.07.09
  • MetaQuotes Software Corp.
  • www.mql5.com
MQL5.community Market provides Expert Advisors developers with the already formed market consisting of thousands of potential customers. This is the best place for selling trading robots and technical indicators!
 
Weedjo:

is it even possible to wirte an EA based on technical indicators, that generates profit over a long period? ...


Yes, possible but it depends on the system itself.
Some people say that forex systems can be profitable with the about 30% up to 80% ROI (or annual profit) for 1 year.
If more than 80 so it becomes non-realistic - those "unrealistic systems" are required to be improved for at least every half a year for example.
But many traders want more profit ... 80% in 1 month ... :)
majority of the forex systems are existing on the market (with profit) for no more than 3 months sorry.

Of course, there are some old classical systems.
And yes, there are some exceptions but those exceptions are based on old classical theories too.

That is why people want to be the first to try some new EA or to download some new indicator.
because it may be too later after some months :)

Just my experience.
 

 

sdsdsd

 

sdsdsdsd 

 
Definitely possible, but not easy.