Once again, about the lokas. - page 52

New comment
 
OnGoing:
That's right. Why do you think that even with SL/TP=10 the probability of accumulating a profit/loss (in the long run) remains the same - 50/50 ?

Everything in the world tends to balance out. So for 10 triggered TPs, one triggered SL is enough to even out, if not lower, your chances for some profit/loss... ;-))
 
OnGoing:
The first answer is more correct. So, because they take the wrong entry points?

I meant random entries, at random.
 
That's right! That's why we have to wait for the signal to be confirmed and on the upswing.
 
Why? We pick indicators that we think give us a signal.
 
Whether formally or justifiably will show the Real score.
 
OnGoing:
Also true, that's what it shows...

It shows what it shows!
 
Mathemat:
Granted it may not be 50/50, but for 99.9% of "systems" it will be as it should be, i.e. 50/50. The whole point is that what the creator of the system thinks is a pattern turns out to be just a futile play on noise or highly dependent things. In other words, the "regularities" of such "systems" do not catch the difference between the process and the martingale.
A very valid point. Find out what the difference is.
 
Mathemat:
Granted, it may not be 50/50, but for 99.9% of "systems" it will be as it should be, i.e. 50/50. The whole point is that what the creator of the system considers a pattern turns out to be just a futile play on noise or highly dependent things. In other words, the "regularities" of such "systems" do not catch the difference between the process and the martingale.


Here the dilemma arises between statistical validity and the limited lifetime of each system. The system designer has a natural desire to increase the number of transactions in tests, because it is the main statistical way to confirm the statistical validity. On the other hand it leads to the fact that the system is revealed late and usually to the end of its life. This is due to the fact that the more time has passed, the more likely it is that the microstructure underlying the statistical advantage will change. Also, those who are losing money on this market inefficiency are gradually getting smarter and there are more people willing to milk them, as many are already paying attention to it.

Therefore, the most important thing is to get the system up and running in a timely manner. Or ways to increase the robustness of the strategy in other ways than just increasing the number of trades on tests.

 
Avals:


There is a dilemma between statistical validity and the limited lifetime of each system. The system trader has a natural desire to increase the number of transactions in tests, because it is the main statistical way to confirm the presence of statistical validity. On the other hand it leads to the fact that the system is revealed late and usually to the end of its life. This is due to the fact that the more time has passed, the more likely it is that the microstructure underlying the statistical advantage will change. Also, those who are losing money on this market inefficiency are gradually getting smarter and there are more people willing to milk them, as many are already paying attention to it.

Therefore, the most important thing is to get the system up and running in a timely manner. Or ways to increase the robustness of the strategy in other ways than just increasing the number of trades in the tests.

Now this is normal. Only I do not understand the place - why on this thread for the feeble-minded?
 
Svinozavr:
Now that's normal. Only I don't understand the place - why on this thread for the faint of heart?

Mathemat pointed out that differences from martingale are not fished out. If you catch it too "well", you'll never catch it)))
1...454647484950515253545556575859...61
New comment