Has anyone created a successful automated trading system? What is your advice? - page 2

 
prostotrader:

You don't have to state what you don't know.

There are systems that are always profitable, but not FOREX, of course!

Well, well... An oligarch who dropped by ???

What systems are "always profitable" ?

Don't be arrogant, let's be on a first-name basis, we're all colleagues here...

 
I agree that there are no permanent systems in place. Another interesting question arises. How can you automate the process of deciding whether a system is working or whether it is time to stop it and reconsider the strategy.
 
Georgiy Merts:

Well, well... An oligarch stopping by ???

What kind of system is "always profitable"?

Don't be arrogant, let's keep it on a need-to-know basis, we're all colleagues here...

Gheorghe, colleagues are those who can do what you do Well. You don't do anything well here, that's why there are no colleagues
 
Vladimir Baskakov:
George, peers are the ones who are good at doing what you are Good at. You don't do anything well here, that's why there are no colleagues

Apart from doing nothing good, throws around unsubstantiated claims.

For all the"Grail" seekers out there!

Constantly, as long as stocks and futures exist, there is a TS that is SUSTAINABLE and GUARANTEED to produce income.

This is trading a stock and futures on that stock.

We buy a stock, sell a futures and at the time the futures expire, we are guaranteed to make a profit.

Added

In the FOREX it is impossible to create a TS which will always bring profit.

If you do manage to do something that works, the VC will not let you make money.

DeC is designed to steal customers, not customers steal from DeC!

 
Vladimir Baskakov:
Georgy, colleagues are those who can do what you do well. You don't do anything well here, that's why there are no colleagues.

Hello.

Are you sure you know what a "colleague" is? That definition doesn't say anything about "good". It just says we have one thing in common. And we have one.

And as for "good"... You're one to talk...

 
prostotrader:

Even if you manage to make something work, the DC won't let you get the money.

I wonder how the DC "won't let you get it"?

That's a trick the brokerage company will do only once. It's the same with any office, and even with any employer... They can always fail to pay. But, as a rule, only once - there will be no more clients for such a brokerage company. That's why any brokerage company has to choose - either to make small profit on spread regularly. Or get big profit on "no profit". But only once.

Not to say that brokerage companies are not interested in hundreds of quid, which an average user of this forum has... Do you want to ruin your reputation for them? The time when forex was "taking off" is long gone... There's no point in fighting over such trifles...

 
Georgiy Merts:

Hello.

Are you sure you know what a "colleague" is? That definition doesn't say anything about "good". It just says we have one thing in common. And we have one.

And as for "good"... You're one to talk...

You're only doing one thing, talking nonsense, you've got colleagues here.
 
Maksim Emeliashin:
I agree that there are no permanent systems in place. Another interesting question arises. How can we automate the process of deciding if the system is working or if it is time to stop it and reconsider the strategy.

The issue of the system not working is solved simply.

Introduce "critical parameters". The history is used to find out the maximum values of these parameters. Then, if at least one of these critical parameters is exceeded during operation, the system is considered to be out of order and needs immediate rework.

The problem is another - how to increase the probability that a system which is working now will work exactly the same for some time to come.

Here, for example, are three of the oldest TSs that I have running in "combat" conditions on a real account:

But who, say, knew in February that TS 440122 would rise so sharply in March and then not "collapse" like many others? Or who can guarantee that the TS 140342 would not fall even more, and would not go to a loss?

 
Georgiy Merts:

The issue of the system not working is solved simply.

Introduce "critical parameters". The history is used to find out the maximum values of these parameters. Then, if at least one of these critical parameters is exceeded during operation, the system is considered to be out of order and needs to be redesigned immediately.

The problem is another - how to increase the probability that a system which is working now will work exactly the same for some time to come.

Here, for example, are three of the oldest TSs that I have running in "combat" conditions on a real account:

But who, say, knew in February that TS 440122 would rise so sharply in March, and then not "collapse" like many others? Or who can guarantee that 140342 TC will not fall even further, and will not go to a loss?

That's the kind of sharply uncertainty that's scary...

 
Georgiy Merts:

The issue of the system not working is solved simply.

Introduce "critical parameters". The history is used to find out the maximum values of these parameters. Then, if at least one of these critical parameters is exceeded during operation, the system is considered to be out of order and needs to be redesigned immediately.

The problem is another - how to increase the probability that a system which is working now will work exactly the same way for some time to come.

Here, for example, are three of the oldest TSs that I have running in "combat" conditions on a real account:

But who, say, knew in February that TS 440122 would rise so sharply in March and then not "collapse" like many others? Or who can guarantee that the TS 140342 would not decrease even more, and would not go to a loss?

I had a client who had 49 robots with different systems on his account and his account was constantly growing. Diversification, so to speak.

Reason: