Non-fitting system - main features - page 15
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
I tried to explain it as I understand it, but I failed a little :o). Imagine, that you have made an aeroplane, but you haven't made the controls a part of the System (a part of the aircraft), but you "took" the controls out of it and gave them to an external optimizer. An engineer on the ground optimized everything for you and you fly, somehow you set up the controls and the whole thing flies with you. You are also told that the "controls" will not change no matter what. Would you get into such a plane? Or let us say you have made control, but the instrument readings are the same, there are many examples.
Optimization is about obtaining parameter values that maximize/minimize the target function (in our case profit), or more spatially - finding extreme values. It only makes sense within the area of constraints that have been imposed. To extend the optimal parameters to a larger area is very presumptuous (and in my understanding just silly), given the specifics of quotes (I mean non-stationarity and high order of the system). Unless there are parameters changing slowly in time, and there are none. And then we start dancing with tambourines - and if we adjust 128 parameters, it will definitely work...
I get the point.
I don't agree that it's presumptuous or stupid to extend it to a larger area.
>> so you can't get out of the house, and there's new inputs and your neuronics can't handle it.
I see the point.
I don't agree that it's presumptuous or stupid to expand to a larger area.
It's not gonna get you out of the house, and there's new inputs and your neuronics can't handle it.
Be careful. A huge number of traders, even very experienced ones, sometimes, and statistically almost always, do not go further than their own backyard.
Optimisation is not about fitting. Optimization is a search for optimal parameters with which the TS will work profitably in the future. If you have found parameters during optimization, at which your TS earns only in the optimization period, and in the future, it is a fitting. Fitting for the optimization period. This is the meaning of the word fitting. If you can not find the parameters of TS, at which it will earn in the future, then you either do not know how to do it, or something needs to change in TS, or something needs to change in the optimized parameters (for example, some of them to fix), or you need to change the period of optimization, or something else....))))
Optimisation is not about fitting. Optimization is a search for optimal parameters with which the TS will work profitably in the future. If you have found parameters during optimization, at which your TS earns only in the optimization period, and in the future, it is a fitting. Fitting for the optimization period. This is the meaning of the word fitting. If you can not find parameters of TS, in which it will earn in the future, then you either do not know how to do it, or something needs to change in TS, or something needs to change in the optimized parameters (for example, some of them to fix), or you need to change the optimization period or something else....))))
Is Martin's optimisation a fit? If yes, why is optimization not a fit?
Optimization is the search for optimal parameters with which the TS will work profitably in the future. If you have found parameters during optimization, at which your TS earns only in the optimization period, and in the future drains, then it is a fitting. Fitting for the optimization period. This is the meaning of the word fitting. If you can not find the TS parameters, in which it will earn in the future, it means you either can not do it, or it needs to change something in TS, or it needs to change something in the optimized parameters (for example, some of them are fixed), or it needs to change the optimization period or something else....))))
Sorry, but you are breaking through an open door I'm not even behind. The phrase is taken out of context. And the post was about something else. Please respond to the substance of it.
Just in case, I'll say it again:
Optimisation is fitting. Simply by the meaning of those words. They are synonyms. There is no negative motive behind it. Just a fact.
For example, what's wrong with adaptability - online optimization, on-the-fly fitting? There would be a sane algorithm on a fundamentally reasonable idea.
So, yes - the main attribute is the need for optimization. But the sabbath is about something else, as I understand it.
How to understand that TS (have stolen it from someone, or what) shows what is logical in it, and was not achieved by an unknown author through optimization. It's a strange thing to do.
No. The subpage is definitely about something else.
Probably about how to understand that the TC you created was made in your right mind, on sober reflection, and is the realization of some idea, not a "Young Alchemist" lab.
No Also somehow...
I'm confused. I'm gonna check out sabj-maker's top post.
Yeah. That's what it's about. How to understand that your idea is correct, and its implementation has the right to be called a stable, profitable working TS.
Then yes - optimization should be used only as a tool for studying the TS for professional suitability. Therefore I propose to return to methods of searching cs through optimization, described by topic starter.
Sorry guys...
But I'll tell you my thought ... Using \ only indicators in a mechanical trading system is like a second-person anecdote ...And using only an indicator in an EA and building any principles only on the indicator and even using an indicator for optimization is like the same anecdote but from the third person ...))))))))))))))
A trading strategy is a function of Equity from the market. For this reason, a trading strategy is a market optimisation.
Evaluating the further behaviour of the function on the basis of its graph cannot give an unambiguous answer. There is therefore no indication of an unsuitable (and fitted) system based on price history analysis.
...
Then it turns out that TA for a comrade is only what is written in so-and-so's book. >> Normal move!
Reread your fiction, comrade, try to read it to the end. And this site has a link to TA, read it too, and don't forget the book (I can send you more, I have 3 gigs of this TA waste paper lying around).
I want to apologise to you for being an "idiot".
Tell me, why do you think I did not call you "xxxxxxxx", "underage xxxxxxxx", or "xxxxxxxxxx", etc.? The only apology I can accept from you is your 3 or 4 years' worth of real-time stats demonstrating how TA works.
to Mathemat
TA doesn't work, not in the traditional or perverted sense. It's enough to look at the DC championships. Why do people like to be so self-deceived, you don't even have to put any effort into it!
TA is about making trading decisions based on market history analysis. The basic tenet of TA is that market history contains all information about future market behaviour. Obviously, this postulate is not true. Otherwise market information (prices and volumes) would contain information about the whole world.
If a strategy uses any complexity of market history analysis to make trading decisions - that strategy is based on TA.
There are only two strategies which are not based on TA:
1. trading decisions are random (this applies to all fundamental analysis decisions).
2. arbitrage.
Is Martin's optimisation a fit? If so, why is optimisation not a fit?
I don't understand, what does Martin have to do with it?
I'm sorry, but you're banging on an open door that I'm also not behind. The phrase was taken out of context. And the post was about something else. I'd appreciate a substantive response.
Just in case I'll repeat it:
I'm sorry, but I didn't see in your post that optimization is the search for optimal parameters for the TS's operation in the future.
But I'll tell you my thought ... I want to use \ only indicators in a mechanical trading system, it's like an anecdote in the second person ... And using only indicator in the Expert Advisor and constructing any principles based only on the indicator and even use for optimization of the indicator, it's like the same anecdote in the third person ... ))))))))))))))
To look at a situation in a different way, that is, as if you see it from the side, from another person - this is a common and widespread trick, which is used everywhere, from science to art .....)))
Why? Where is this information coming from? For example, my EA on the champ used TA exclusively, but in a somewhat unconventional interpretation....))))