A great book on testing and optimisation - page 10

 
autoforex >>

One last question - who will the system really work for? Me or Vasya?

I highly doubt that Vasya, such a cool mathematician, will be able to justify with his calculations the best of what you have done. You will have different results, i.e. different parameters. Or do you too, like some others, think that statistics is pseudoscience?

2 FOXXXi: has anyone already proven that the market is a martingale? I don't believe so. If you yourself say that perpetual systems exist, then consequently they are bound to use its non-martingale.

 
LeoV >> :

I don't mind what you say at all. On the contrary, I'm even for it. But can you say something more sensible other than blaming everything? )))) Or if you can't, then can you at least confirm your words with some real life or monitoring? So that everyone can really get a feel for what you're saying. Not with screenshots. I don't need screenshots. ))))

I don't follow PAMMs, but you must be aware of it. Their results confirm the coolness of my words or not? I usually go to an adequate investment company and there I can prove why the system will work in the future and get the money to manage.It should not just work, and ahigeno and work steadily, you must be in the numbers to show the annual yield in percentage terms per unit of an asset. There three skins down, they do not need optimization and steep "self-adaptation" neural networks for the previously unpredictable process.

I made an experiment: I called to investment companies and said that I had a profitable system based on TA - some companies immediately stopped talking to me, others had enough patience to listen to my beliefs, and that's the end of it.

You have a strange logic: if I'll tell you about it, prove it, show you millions from my pocket and you'll say "sorry, it's not working, you were right". If I won't do it, then TA and neural networks should be trained.

What kind of monitoring can we talk about in my case, it's a different mentality for me.

 
Mathemat >> :

2 FOXXXi: Has anyone already proved that the market is martingale? I don't believe it. If you yourself say that perpetual systems exist, then consequently they are bound to use its non-martingale.

And what is a martingale? It is simply a random process with zero mathematical expectation, the best prediction of which is the REAL state. A Wiener process is a martingale, and a Wiener process is a model of Brownian motion, i.e. the random wandering of particles.I've written about it myself and posted a picture somewhere with time inefficiency of real series with statistical significance. But if we work on it, we will fly like plywood over Paris, because we don't know when it ends, the payoff is low, but the risks are enormous.

What is there to prove. take the same euro/dollar. zigzag, not the standard one, but the one, which the beam can be set in pips, and you may run it on minutes till the blue moon, optimize parameters. if you find the right parameter, it will be big, and it means you need more data.

 
FOXXXi >> :

Take a Euro/Dollar Zigzag, not the standard one, but the one, which can be set in pips, and run it on a minute until it's blue in the face, optimise the parameters.

>> No way, I do not look at such difficult indicators for a long time. I do not look at them either :)

 
FOXXXi писал(а) >>

I do not follow PAMMs, but you might know about it. Their results confirm the coolness of my words or not? I usually go to an adequate investment company and there I can prove why the system will work in the future, and get moneyshku in management.It should not just work, and ahigenic and stable work, you must be able to show the numbers in the annual percentage yield per unit of an asset. There three skins down, they do not need optimization and steep "self-adaptation" of neural networks in a previously unpredictable process.

I made an experiment: I called investment companies and told them that I have a profitable system based on TA - some companies immediately stopped talking to me, others had enough patience to listen to my opinion and that was the end of it.

You have a strange logic: if I'll explain, prove, show millions from my pocket, then you'll say "sorry, all this stuff doesn't work, you were right". If I don't do it, then TA and neural networks should be able to prepare.

And in general, what kind of monitoring in my case can we talk about, it's some other mentality for me.

There's no reason not to believe you. But there's also no reason to believe you. So you give an example of something more real, not a verbal bazaar on the topic of "How fucking cool I am". Words are wind, air, they are unreal. No one is responsible for them, on the internet. So be more real and you'll get kudos and respect.....)))))

 
LeoV >> :

There is no reason not to believe you. But there is also no reason to believe you. So you give an example of something more real, not a verbal bazaar. Words are wind, air, they are unreal. No one is responsible for them, on the internet. So be more real and you'll get kudos and respect.....)))))

I repeat, I have laid out the basic material, the model that won the Nobel Prize. With this model you can get a stationary series. You have to take this model, study it if you do not know it and apply it to the market. This is an extensive research analysis, this work. The goal is to find this proess(es) using the presented model.You want to disprove the Nobel Prize, go and do it, do not extort the system. You have nothing in favour of TA, that it works and no argument for the inoperability of my system. I mean my words are not wind or air and they are real. I do not hide the basis for my system. Only the instruments found / traded are subject to conspiracy. The maximum that you will get, is screenshots of these processes, of course without specifying the instruments. If they are needed, I will lay them out, I am kind today.

 
FOXXXi >> :

>> I'm not hiding what my system is based on.

Let me guess?

The unsophisticated conversion is called cointegration and the TS is based on statarbitrage?

I trade it myself and I have about 200 instruments.

 
goldtrader >> :

Let me guess?

The conversion is called cointegration and the TS is based on statarbitrage?

I'm trading it myself and the instruments are under 200 somewhere.

I am telling you, there are such people here, they are just talking rubbish sometimes. I am writing, they do not refute a word, I have air in my words. Explain to LeoV more really what's what. Maybe they together with Reshetov and Matemat are talking rubbish, if not - it's clinics.

 
FOXXXi >> :

I am telling you, there are such people here, they are just talking rubbish sometimes. I am writing, they do not refute a word, I have air in my words. Explain to LeoV more really what is what. Maybe they together with Reshetov and Matemat are talking rubbish, if not - it is clinics.

Well, you're wrong, dad. Just because one thing works does not mean that the other does not or cannot work.

All the more the audience here is specific - sharpened for Forex, this approach is unconventional for them. )))

 
goldtrader >> :

Well, you're wrong, dad. Just because one thing works doesn't mean the other doesn't or can't work.

I'm not going to rehash that "other".

Reason: