How to spot fake EAs from the market place? - page 2

New comment
 
Camilo Mora:

Recently, I watched a video of a person coding an EA that took historical data, offset it by 1 candle and loaded it into an EA that tailor positions based on whether the prior candle went up or down. The result was an EA with amazing performance and very low DD. The point it was trying to make was how easy is to bias the results of the backtesting. Can anyone think of a method by which this type of situation can be identified?

The only I can think of is the use of the live signal, which has its own biases as in most cases it is very recent, and in some other cases authors load several and remove the loser ones over time. The facility by which current EA performance indicators can be manipulated, leaves people vulnerable to scams while rendering useless the work by honorable developers.

While one can buy based on trust on the developer, this will make it superhard for new developers to enter the market.

Create a custom symbol with inverted data and test the ea there .

The coder has to know which symbol their ea is running on so they'll either look for the pair's string existing in the symbol name , or , use the SymbolInfo to get the currencies involved.

However : If someone else has trained a neural net to trade that symbol in their EA , they also need to know the symbol and you will also reject them with this process because you are flipping the data.

 
Lorentzos Roussos #:

Create a custom symbol with inverted data and test the ea there .

The coder has to know which symbol their ea is running on so they'll either look for the pair's string existing in the symbol name , or , use the SymbolInfo to get the currencies involved.

However : If someone else has trained a neural net to trade that symbol in their EA , they also need to know the symbol and you will also reject them with this process because you are flipping the data.

One issue with changing the data, is that if the EA is a predictive model, then by changing the  data it will render it useless. Think of it like this... consider a thermometer, that turns red when it is hot or blue when cold...if you change the environment such that when it is suppose to be cold is hot and viceversa... then the thermometer will give us a wrong/reversed reading. The idea of "non-gambling" trading is that there are patterns in the data, that one can identify, via different approcahes... changing the data, I fear destroys those patterns.
 

Hi,


Intersting topic as I am also concerned about this question. If the ea has been released/updated few months ago and the backtest afterwhile runs fine does it mean the backtest is legit ? I mean is there another way for the creator to cheat like reading the next bar values before the backtester module (without DLL access) ?

Thanks

12
New comment