Mechanisation of optimal parameter selection. Finding a common denominator. - page 8

New comment
 

lasso:

......

Let's post here the results of optimization of an EA on MA included in MT4 delivery, (or any other, no problem)

and you try to select a "good set".

....
There is one small but. You can't optimise the shit. I mean you can, but it's useless. You need a "fruitful opening idea" ))
 
lasso:

Constructive:

let's post here the results of the optimization of the EA on the MA included in the MT4 delivery, (or any other, no problem)

and you try to select a "good set".


That's not constructive, that's overblowing) In addition to a good set, there has to be a meaningful robust system and the mashka from the MT4 delivery is not one of them.
 
Avals:
No, there was no desperation - there was reinvestment))) But I agree that this is evidence of a personal nature. Just any way you look at it, it still comes down to backtesting on history. Even a real test is already a story)) And you'll encounter statistics anyway, if you want a statistical advantage rather than a guessing game :)

I would say that this is proof of a one-off nature. There was a similar case on pamm A..ri for example, managed to raise 11000% in 3 months as I recall.

As for stat. advantage, it should be there, but not based on backtests.

 
OnGoing:

I would say that this is proof of a one-off nature. There was a similar case on pamm A..ri for example, managed to raise 11000% in 3 months as I recall.

As for stat. advantage, it should be there, but not on the basis of backtests.


then it's just an advantage - not statistical : )
 
Avals:

then it's just an advantage - not statistical : )
Inside ))
 
OnGoing:

I would say that this is evidence of a one-off nature. There was a similar case

)))

What's the problem with admitting it as evidence?

 
lasso:

IMHO.

1) The TS to be optimised should work with a constant lot.

2) Not more than one position at a time, otherwise it is in fact already two TS or more.

1) This condition is not suitable for all the TS. In some TS, changing of the order volume may not be a part of the MM.

2) It doesn't have to be two or more TSs. Example: We obtain the entry signal on each bar but the orders have TP and SL, thus several orders will be in the market simultaneously within one TS.

//////////////

I suggest the following, as a criterion for parameter selection among the optimized parameters we have already got

a) if the volume of orders is constant then:

K=Prr/Sub.ser.

where:

Ppr - percentage of profitable trades out of the total number; Sub.serv - maximum series of losing trades

b) if the volume of orders is "floating" (see point 1), then

K=Av/Sub.ser.

where:

Av - average value of profitable trades; Sub.ser - maximum series of losing trades


Naturally, the larger K, the better.

 
Avals:

then it's just an advantage - not statistical : )
It's not that simple, my friend) but about that later...
 
OnGoing:
It's not that simple, my friend) but about that later...

When ?
 
Mischek:

When ?
When the cancer is on the mountain...
123456789101112131415...19
New comment