Where is the line between fitting and actual patterns? - page 3
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
And the most reliable :o)... In fact I tried to trade on the real without looking at the monitor or turning it off by looking at an upside-down monitor or reflecting in a mirror.) I traded using signals of my fellow traders who were steadily losing money, i.e. I reversed them. I have never tried to buy my forex but I have not managed to buy it. But the truth is that when the market is more than psychological bar (in my case it was more than five standard), then for some reason it starts the reverse process... That is, my friend starts to actively earn :o). I traded on the signals of a fortuneteller using Tarot... Not knowing what it is about three weeks in a row said what colour will be the next "stick" (the daily candle index of the Frankfurt Stock Exchange) and consistently guessed... But when my friends and I bet some money on this prediction and waited all day for this bl day to "repaint", the four of us lost $32,000 for the day :o)... So it's not all that simple guys. And if you think history is the stuff to write a system, you're screwed :o)... Let's write an EA that's supposed to drain in an ugly way! Badly, hopelessly, badly... And that's when you can look for edges.
interesting approach, i heard people use a pendulum right on top of the terminal at a beech reversal - they ask the pendulum where the next moves will be and it answers them - it seems to work...
And it works. I drew a graph of the Euro on a tape measure about two years ago using a pendulum. You will laugh, but the picture a month ahead was very similar. Not much shooting quotes, but the dates and the sense of what was going on was guessed quite well. I tried to trade smaller lots later. During the day I spontaneously approached the machine and asked what exactly to do now. You may try it if you have time between testing your EAs:o). A sheet of A4.Two lines in pencil.In the middle is vertical, in the middle is horizontal.That is cross.Vertical line from you with the signature "YES".Horizontal line with the signature "NO".I am doing an armrest pose, a sheet under your index finger, which is an elbow on the table. On your index finger, there is an object which works as a pendulum. I had a simple sinker from a bottom fishing tackle :o) on twine of about 30 cm length with a loop under your finger. That's it :o). Drive everybody away from you, go away into a bunker, pour concrete in the entrance, if possible, stop thinking about borsch, chicks and TV, but actively ask a question in your mind, what shall I do now and hold a pendulum over the crossed lines, having calmed it from dangling... It is necessary to ask a specific, as precise a question as possible without options and watch the reaction of the device.For example - Should I buy a pair of fuetjen now? If it starts to wiggle on the vertical "YES" line - buy, if on the horizontal "NO" - don't buy, but don't sell either. If it starts to twist around the cross in any direction - don't do anything.
That's artlessly spend their leisure mid-term traders :o).
p.s. Respect, akhtung, atention, attention ... I have not reached this level of enlightenment yet... If I do, they say it's because the real money does not come that way. And if I reach it, they say that the money is of no use...
And it works. I drew the EUR chart on a tape measure about two years ago using a pendulum. You'll laugh, but the picture a month ahead was very similar. Not much shooting quotes, but the dates and the sense of what was going on was guessed quite well. I tried to trade smaller lots later. During the day I spontaneously approached the machine and asked what exactly should I do now. You may try it if you have some time between testing your EAs :o). A sheet of A4.Two lines in pencil.In the middle is vertical, in the middle is horizontal.That is cross.Vertical line from you with the signature "YES".Horizontal line with the signature "NO".I am doing an armrest pose, a sheet under your index finger, which is an elbow on the table. On your index finger, there is an object which works as a pendulum. I had a simple sinker from a bottom fishing tackle :o) on twine of about 30 cm length with a loop under your finger. That's it :o). Drive everybody away from you, go away into a bunker, pour concrete in the entrance, if possible, stop thinking about borsch, chicks and TV, but actively ask a question in your mind, what shall I do now and hold a pendulum over the crossed lines, having calmed it from dangling... It is necessary to ask a specific, as precise a question as possible without options and watch the reaction of the device.For example - Should I buy a pair of fujien now? If it starts to wiggle on a vertical "YES" line - buy, if on tgrizonal "NO" - do not buy, but also do not sell. If it starts to twist around the cross in any direction - do nothing.
That's how ingenuous mid-term traders spend their leisure time :o)...
p.s. Respect, ahtung, atention, attention ... I have not reached this level of enlightenment yet. And if I reach it, they say that the money is of no use...
...You forgot to add, first of all, you have to tune the pendulum... Something like: "Pendulum, pendulum, answer me how you will answer "yes" - to my question (let's say the pendulum answered by rotating clockwise, then - ask the pendulum - how it will answer "no" to the question, before that you should ask the pendulum at all about its readiness to work and answer questions...
Well, go ahead... You've touched on a "lively" subject. My first higher education is technical, I graduated from the Faculty of Information Technology at the State Technical University in 1999. University - so there we really taught the subject - bioenergoinformatics - i.e. aura, chakras, pendulums, etc.. - It was interesting all this, i.e. we (a subgroup) at seminar session sit each at the desk, really with pendulums in hands above tables - we communicate, we ask questions, we receive answers... It's funny. I have not yet applied TC as an option, I am digging in other directions, though anything is possible... The main thing is the right approach...
Perhaps the signals filtered by coffee grounds and phases of the moon ... (as an option) - :- ))
P.S. I'll consider the possibility of dedicating part of my deposit to research in this direction...
P.P.S. I think there will be a worthy continuation - Fundament, TA, spectral analysis, neural network, ... :-))), why not, it is possible to try...
I don't understand, how to trade, if not by the TS? By randomness? This is also a TS )))).
I don't understand the question, but trading is like adjusting to the current market, and if it changes, adjusting and trading again.
And not adjusting is only when you yourself affect the market.
All the rest are fitting according to the patterns found. And they don't last forever. Some last for months, some last for years.
Where is the line between fitting and actual patterns?
The line can be calculated by comparing results between Sample and OOS
Many neural network packages have a very handy way of separating the sample into training and test samples. It is very easy to detect a fringe. If there is no improvement at all on the test sample then it is a naked fit. I.e. the inputs are non-kosher and the TS can be discarded. In other cases we look until the moment when the results continue to improve on the training sample, and no longer improve on the test sample - this is that very edge.
Another thing is that the separation of historical data on the optimized and test sample is not implemented in the MT* tester; therefore it is almost impossible to detect the border in this situation during optimization.
Experimented with TS operating with probabilities I noticed that the fit appears when some historical events have zero probability. Any event that occurred in history must have a probability higher than 0.
A facet can be calculated by comparing results between Sample and OOS
Many neural network packages have a very convenient way of separating the sample into a training sample and a test sample. The boundary is very easy to identify. If there is no improvement at all on the test sample then it is a naked fit. I.e. the inputs are non-kosher and the TS can be discarded. In other cases we look till the moment when results on the training sample continue improving, and on the test sample they don't - it is that very edge.
Another thing is that division of historical data into a sample to be optimized and a test sample is not implemented in the MT* tester, therefore it is almost impossible to detect the difference in this case during optimization.
Experimented with TS operating with probabilities I noticed that the fit appears when some historical events have zero probability. Any event that occurred in history must have a probability higher than 0.
That's right. There is simply no better way to check how well optimised the TS is (not fitted).
One problem remains. The relevance of such optimisation lags exactly by the value of OOS. Well yes, someone will say, all that remains is to optimise the TC up to the present moment.
Booga-ha! And there's no test OOS period for comparison.
One can't help but think that we must choose the OOS size as small as possible to increase the optimization relevance. But there is a problem here as well - the smaller OOS, the greater the probability that the real-time MT will fail.
It is a stick with two ends, so to say. Or even three ends.
joo:
But therein lies the problem - the lower the OOS, the more likely it is that the TS in real life will drain.
The answer is wrong. Just when training NS I take the OOS period less than the period of the training sample. BP is non-stationary and if we do the opposite, we just get a fit to a short sample and a very doubtful result on OOS.
Simply put, for nonstationary VR, the shorter is the period of OOS in comparison with Sample, the less probable that the market has time to change for such a time, i.e. some part of regularities found in the history can remain. And the larger is the Sample period, the lower is the probability of adjustment. But do not go overboard with Sample, because very large period can give the adjustment for already inactive regularities, which took place in the distant past and unlikely to repeat in the future.
You forgot to add, first of all you need to adjust the pendulum. Something like: "Pendulum, pendulum, answer me how you will answer "yes" - to my question (let's say the pendulum answered by spinning on the hour hand, then - ask the pendulum - how it will answer "no" to the question, before that you should generally ask the pendulum about its readiness to work and answer questions...
Well, go ahead... You've touched on a lively subject. I have a first degree - technical, '99, the Faculty of Information Technology of the State Technical University - and there we really taught a subject - bioenergoinformatics - i.e. aura, chakras, and chakras. University - so there we were really taught a subject - bioenergoinformatics - i.e. aura, chakras, pendulums, etc.. - It was interesting all this, i.e. we (a subgroup) at seminar session sit each at the desk, really with pendulums in hands above tables - we communicate, we ask questions, we receive answers... It's funny. I have not yet applied TC as an option, I am digging in other directions, though anything is possible... The main thing is the right approach...
Perhaps the signals filtered by coffee grounds and phases of the moon ... (as an option) - :- ))
P.S. I'll consider the possibility of dedicating part of my deposit to research in this direction...
P.P.S. I think there will be a worthy continuation - Fundament, TA, spectral analysis, neural network, ... :-))), why not, it is possible to try...
In fact, as far as I know, statistics for people employed in different areas looks the same with a ratio of approximately 62%... We are fully "immersed" in that figure. That is, such a ratio is necessary for the survival of the species, the non-powerful should be less than half. In the CIS we are now seeing a skew towards the "non-powerful", especially in Russia and Ukraine.They do not want to work and are trading the narrow-eyed crap and as a consequence the trend is clearly downwards. The situation with gambling activities (betting in particular in the form of stock trading) is not natural for people. :о)
And in a short time the DT as a business will cease to exist :o)
Trading history has a lot of artificial corrections and is therefore unnatural, but we can fit almost any algorithm to it.The only thing is that it will be successful for a certain time period and it's good if you escape with the profit in time :o)
.... Exactly 5/95% not for the better....
Where did you get these statistics from, please?
Same place as the pendulum...