"The 'perfect' trading system - page 25

 
VictorArt писал(а) >>

And rightly so.

Why do those who don't trade need advisors? :)

We'll see what excuse you come up with for why the "profit period" never started.

 
paukas >> :

We'll see what excuse you come up with for why the "profit period" never started.


Are you talking about PAMM?

I can immediately give you an "excuse for one of the future options": "given a large number of adaptive trading robots in one account, somewhere we made a mistake with the balance between profit and loss - losses turned out to be greater than profits" :)

 
VictorArt писал(а) >>

Are you talking about PAMM?

I can immediately give you the "excuse for one of the future options": "given the large number of adaptive trading robots in one account, somewhere I made a mistake with the balance between profit and loss - losses turned out to be greater than profits" :)

I can also recommend:

- the stupid operator interfered.

- the market behaved inadequately.

- the broker did not allow the robots to work correctly.

- Not enough money - if I had a million, I would have shown everyone.

etc. etc. :)

 
paukas >> :

I can also recommend:

- the stupid operator was interfering.

- the market behaved inadequately

- the broker did not allow the robots to work correctly

- not enough money, if they had a million - then they would show everyone

etc. etc. :)


No good, as all such excuses are already in use by others :)
 
VictorArt писал(а) >>

Thank you for your answers. Unfortunately, what you said did not satisfy me in any aspect. Everything you have written is understandable, but it is nothing more than general phrases.

I am not aware of any work on finmarkets that does not fall into the category of "near-scientific jibber jabber" :)

FIN markets are not science yet.

Have you never heard the name Shiryaev? And have you never held in your hands his monograph "Fundamentals of Stochastic Financial Mathematics"? But this is just an example. I personally have nothing against scientific jibber jabber, all the more so here on this forum. But when someone ventures to speak about the "General Theory of Trading", one wants to see the origin of this theory, its apparatus, methods, and most importantly - the results (not to be confused with the results of trading). However, except for a phrase that sounds like a good wish, in fact there is nothing else.

The financial markets are definitely not a science and they will never be. It is the realm of real finance, where money and equated assets are circulated.

The term 'synchronise' is a standard, standard synchronisation of two processes.

For example, what happens if the development of NF is "predicted" so that it corresponds as closely as possible to FR? Over time, due to the accumulation of prediction errors, the divergence between the two functions may reach such large values that the prediction will lose all meaning - the functions will become completely asynchronous, i.e. will exist on their own.

This is why synchronisation is needed.

If you have your own function, you don't need to "predict its evolution". It exists on the whole definition area from the moment it is defined, so "predicting development" in this situation sounds like guessing by coffee grounds.

In general, in mathematics, in the theory of operators, there is such a term - the eigenfunction. And you use it in the sense of "I invented it myself, so it is my own function". Ridiculous. Especially funny is the process of predicting it after you have invented it yourself.

What you call synchronisation is called in mathematics an approximation of a market function to some model function. And "predicting development" in the language of mathematics is called extrapolation.

How you will synchronize it is not so important, as this is the level of algorithm. For example, in the adaptive EA example, the synchronization is done through a stop loss. Stop Loss triggered - changed the direction.

The search for a high-quality approximation of a market function is directly or indirectly engaged in everyone who is trying to write trading robots. Qualitative means that the extrapolation of the model function into the future should approximate the market function quite well and on as large an area as possible. So you haven't said anything new here.

If the model function is defined, then the problem of fitting (in your language - synchronization), i.e. choice of the optimal parameters, is purely technical and therefore it is really solved at the level of the algorithm. Another matter is the choice of the model (in your language - your own) function itself. However, it turns out that OTT is powerless here and cannot help the trader. Good theory!

However, I am probably wrong. You are saying that "the NF can be anything". Have you ever tried to trade with any function? I don't think so. That's a pity. Then you wouldn't be making such baseless and irresponsible statements.

And this pack of platitudes in general is hilarious:

The more stable the eigenfunction is, the more stable the result will be.

The better she lets herself get in sync, the more accurate the sync will be.

The more accurate the synchronisation - the greater the profit, i.e. if the NF and FR are fully synchronised, it will mean no losses.

.

In general, I do not understand what else there is to describe in more detail - everything is clear as it is :)

Yes, you're right, everything is clear. I don't know how your digital brain works there, but the general theory is clear - there is none.

When someone comes up with an even more general trading theory than OTT, then I'll think about changing the name to something more modest :)

I've come up with one. Beg your pardon: "To make a profit, you should only open profitable trades".

I've decided to call it Universal Generalisation of General Trading Theory. Or GETT. I think it's nice. :-)

 
Yurixx >>:Впрочем, я наверное неправ. Вы ведь утверждаете что "СФ может быть любой". А вы не пробовали торговать с любой функцией ? Думаю, что нет. А жаль. Тогда бы вы не делали такие безосновательные и безответственные заявления.

PRNG is a random function, i.e. any function :)

Quite good results are obtained - see traders' contests.

In general, you are once again trying to apply mathematics to the market, that's all.

 
Yurixx >> :

If the model function is defined, then the issue of fitting (in your language - synchronisation), i.e. choosing the optimum parameters, is purely technical and therefore really solved at the level of the algorithm. Another matter is the choice of the model (in your language - your own) function itself. However, it turns out that OTT is powerless here and cannot help the trader. Good theory!

What can I tell you?

If you confuse synchronization with fitting...

One more thing. A proprietary function doesn't model anything or try to model anything, that's why it's proprietary, not a model function.

 
Yurixx >>:То, что вы называете синхронизацией, в математике называется аппроксимацией рыночной функции некоторой модельной. А "предсказание развития" на языке математики называется экстраполяция.

In general, it's all clear here - you twisted it in your own way without understanding it, and then you found it funny - what nonsense it turned out :)

I'm laughing too.

 
Yurixx писал(а) >>

.... "To make a profit, you should only open profitable trades".

....

Fair enough!

 
VictorArt писал(а) >>

I am not aware of any work on financial markets that does not fall into the category of "near-scientific jibber jabber" :)

FIN markets are not science yet.

Sorry, have you not heard of the Black-Scholes formula and generalized autoregressive conditional heteroskedasticity models?

Or is that "sci-fi talk" to you as well?