Stereo Neuro Net - page 10
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
here are the numbers instead of the "crap", and the trend/flat thread is good, but with a link to price, and that's like looking for a line between walking and jogging
Rosh, by the way, today he uses consolidation instead of flat)) i like it better than flat-flat.
Nothing against "consolidation" ;)
And I'm just using this as an example of constructive conversation.
I came to the forum for once and there's practically nothing but "woof-woof"...
Can you elaborate on the difference? the difference with what? margin means credit, more risk
the market has not moved, the deposit would not have moved either, but the margin prevents it, do you want to live in a static world?)
I'm afraid this topic is going to get philosophical about personal value.
You can. In margin trading, there is no real delivery of currency. You cannot simply buy (as an example) an Australian dollar and put it in your account (as well as other currencies). In margin trading you may only have an open position, i.e. against something (another currency) e.g. buy AUD/JPY, sell GBP/AUD, etc. So there is no way to create a basket of currencies so that it "doesn't move anywhere" in margin trading.
not so sad))
in 'MathRand() function' even random entries give a profit of 840 runs out of 1000 on the period 1999-2009
this is being achieved due to trend prevailing over flat in proportion of about 5 to 1 on large timeframes
I'm afraid to disappoint you, but you are wrong in this formulation. At least you need a trend and flat criterion on your part, in which your 5/1 proportion is met.
Coaster confirmed it with charts on small timeframes too, not with the same effectiveness, of course, but I don't know what conclusions he drew from it.
I looked through the 'MathRand() function' branch (no one interfered) with similar conclusions. Where did you see profitability with MO in the range of 0.5-1.0 points with no commission?
here's the figures instead of 'coddling' and the trend/flat thread is good, but with a link to price, and that's like looking for the line between walking and running
Would love to see a non-price criterion.
Rosh used consolidation instead of flat today) I like it better than flat-flat.
Consolidation, as a rule, is a contraction of price (characterized by a triangle shape) and it is hardly possible to consider it a flat, because the range value changes. Consolidation examples
When formalising trend/float in terms of profit taking (rather than in terms of physics or philosophy), neither trend nor flat is profitable, as it is present in the market for about the same amount of time (+/- 2%). Of course, this is true only for the method used to separate the trend from the flat.
Let me clarify it a bit. Not an equal amount of time, but an equal amount of points. By time (number of bars) there is a slight prevalence of flat areas over trend areas. Quantitative characteristics and statistical data on some GPs are given on the 16th page of the Trendo-Flat thread.
I came to the forum for once and there's practically nothing but "woof-woof"...
"But he's all "moo" and "moo", And why, I don't understand! - Hang up the phone, please!"
I'm afraid to disappoint you, but you're wrong in this formulation. At least you need a criterion of trend and flat, when the proportion 5/1 you specified is fulfilled.
I looked through the good thread (no one interfered) with similar conclusions. Where did you see profitability with MO in the range of 0.5-1.0 points with no commission?
Would love to see a non-price criterion.
Consolidation is usually called price contraction (characterized by a triangle shape) and can hardly be attributed to flat, because the range value changes. An example of consolidations
(Your tenacity amazes me, I want to shake it off and forget it as a bad dream, but I can't))) and you have nothing special to answer with, you just said "no flat", composter says about the equal proportion of the flat / trend, you mean the small-time predominance of the flat, and I bent my words 5 to 1 in favor of the trend
It's unlikely that my words will change anyone's point of view, so I'll be brief
1. I measure the market dynamics using several methods, since the object of research is unclear and blurred (flat/trend), while its significance for the TS is paramount, I use eyes, ears, sense of smell, intuition, sense of touch, and other instruments
2. anyway in different threads i mentioned all these ways of measuring, for example in 'MathRand() function' there are two ways, one is trend EA which is losing 10 months out of 60, second is the number of runs of random trend EA - 16% are losing, hence the ratio 1 to 5, these are indirectly price related measuring ways, there is also the mentioned in this thread "i.e. three indirect methods, these methods have an important unit of measure for flat/break - average deal time.
3. There are 2 methods for measuring the price directly, one is described in the topic "predicting a price is possible" in the sentence "mathematics is how to evaluate the quality of the forecast, but it is also applicable to the evaluation of a flat/trend, the second in this topic, like a flat - is the price return, elementary calculation of the number of returns at a given time interval, you can also add here the method described in your topic, but I have not yet read it
4. All price methods are connected with history, and that means they are practically useless, so it is clear why the result is flat/trend=50/50, the simple reason is that they do not take into account the indirect methods related to trade operations and described in item 2, In practice it turns out that there is no alternation flat-trend-flat-trend, but in practice the trend includes at some parts what the unbiased price methods would call flat, and then the question appears: should we exclude these parts from flat calculation or not? If we exclude, then we will come to the ratio 5 to 1, if not, then the ratio will be less - 5 to 3. Indirect methods also deal with history, but they show statistical advantage or weakness
5. Even with simple price methods the prevalence of a trend is obtained, but not of a flat, if you want I can describe it, but as I don't feel enthusiastic about price methods, I think that to describe it means to enter into unnecessary disputes
6. The most interesting one, of course, is the methodology not related to the price in any way, that's why it can make predictions. All other described methods are just needed to calibrate this method. It is well known to those who trade with long handles, they know for instance that there will be slack for a while, and if there is a storm, it will not stop immediately, if the trend stopped, it does not mean that it has finished, etc. Currently I formalize these principles
7. Personally, for me the notions of flat and trend give much to give up; I have rather found not quantitative, but qualitative aspects of these notions, like in the video clip on the first page where two classes of phenomena emerge from smoothly changing states - they are like heaven and earth
I've looked at the good thread (no one interfered) with similar conclusions. Where did you see profitability with MO in the range of 0.5-1.0 points with no commission?
and do you have to add spread, requotes and name you can't pronounce? what does it have to do with flat and trend?
As a rule, the consolidation is a price compression (characterized by a triangle shape), and it can hardly be considered a flat, because the range value changes. Consolidation examples
thanks, i will use the correct meaning.
i don't want to write too much, i got a whole message to the world, i need to restrain myself)
so I'll be brief [...]
6. the most interesting of course is the methodology not related to the price in any way, so it can predict, all other described methods are only needed to calibrate this method, it is well known to those who trade long handles, they know for example that there will be a lull for a while, and if there is a storm, it will not stop immediately, if the trend has stopped, it does not mean that it has ended, etc.
Well yes, that's the part where you're particularly brief, just brilliantly brief.
About the meteorological phenomena: I'm not the only fool who tries to formalize them. But I still can't understand how this methodology has nothing to do with price...
By and large, apart from the first post - all are leftists.
Dear komposter,
the first post is just a cartoon demo reel on neural networks, and there's NOTHING else,
Type in: neural network demos on Yandex...
are forum visitors really stupid enough to fall for these cartoons?
About weather phenomena: I'm not the only fool who tries to formalize them. But I still can't understand how this methodology has nothing to do with the price...
of course weather is related to price, for example if you assign different coloured candles to 1 and -1 and add them up over the year, you get the following picture, which shows that the higher the p(B)parameter pressure, the less candles in the residue and vice versa
of course the weather is related to the price, for example if we assign 1 and -1 to candles of different colours and add them up for a year, we get the following picture, which shows that the higher the p(B)parameter pressure, the less candles are in the residual and vice versa
interesting study, but how will this methodology work on minute M1 charts?