The market is a controlled dynamic system. - page 305
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
the widget doesn't work.
To insert a widget, you need to be the administrator of the resource. This is possible on your own website or, for example, on a WordPress-based website (if an HTML extension is installed).
To insert a widget, you have to be the administrator of the resource. This is possible on your own site or on a WordPress-based site, for example (if you install the HTML extension).
It's a strange approach -- the trader has to be the administrator of the resource, or negotiate with the resource's administrator in order to insert the widget... With that approach, the widget is useless.
You will remember what the widget is created for and for whom, what the function of the widget is.
Widgets can be inserted without any problems, for example.
Think about the fact that by taking the differences as an input, you are eliminating the trend component from consideration, and therefore you end up with exactly rubbish,
However, this is the input signal. The marketplace is just the adder (integrator) of this input signal. As you can see from the picture, there is nothing obvious (trends) in it. Otherwise there would be oscillations of noise around zero. That is, all movements are deep under the noise.
Now take Brownian motion - initially there are no trends, only noise and nothing more. If we add them up, we will get trends, flat patterns and an indistinguishable picture from the market. Everything can be done in any "Matrix" in 15 minutes, if you wish.
Here is a picture of Brownian motion. It's just a case). Market time series have most of the properties of Brownian motion series.
https://ru.wikipedia.org/wiki/Винеровский_процесс
The normal distribution is easily reduced to the real market distribution, then the pictures will be indistinguishable at all. In other words, just a random process generated the time series equivalent to the market ones which we analyze and try to find some regularities in them.
Think about the fact that by taking differences as an entry, you eliminate the trend component from consideration, and therefore you end up with exactly rubbish,
It's not rubbish - it's volatility, trades beautifully.
The picture shows large deviations from three sigmas - these are just the thick tails and it is the behaviour of the trading system when returning to a stationary state that is fundamental. World crises are just thick tails: seldom but quickly.
And what is a trend if we cannot distinguish a reversal from a correction?
It is not rubbish - it is volatility, traded perfectly.
The picture shows large deviations from three sigmas - these are just the thick tails and it is the behaviour of the trading system when returning to a stationary state that is fundamental. World crises are just thick tails: seldom but quickly.
And what is a trend if we cannot distinguish a reversal from a correction?
San Sanych, trade in the channel, if you know how, defend the honour of the 101st institute :)
It is not rubbish - it is volatility, traded perfectly.
The picture shows large deviations from three sigmas - these are just the thick tails and it is the behaviour of the trading system when returning to a stationary state that is fundamental. World crises are just thick tails: seldom but quickly.
And what is a trend if we cannot distinguish a reversal from a correction?
Exactly, rather than trading an adder (market) output that lies deep under the noise and indistinguishable from Brownian motion, it is much more efficient and easier to trade statistically stable input rubbish. And it is more efficient to trade not the unlikely tails, but the centre of the distribution, which is about 50-60 p (for futures) at the high. And the tails will come and fit there on their own, without any prediction.
And why not all at once? Let's take an oblique t-distribution and go....
That's right, instead of trading the adder (market) output that lies deep below the noise and is indistinguishable from the Brownian motion, it's much more efficient and easier to trade statistically stable input rubbish. And it's more efficient to trade not the unlikely tails, but the centre of the distribution, which is about 50-60 p (for futures) at the high. And the tails will come and fit there on their own, without any forecasting.
Well... If for you the market movements are"indistinguishable from Brownian motion" andit's "easier for youto trade statistically stable input rubbish", then so be it.
But don't bring up"tails", let alone that"tails will come by themselves and fit in".
And about"predicting" - this is not the place. There was a huge thread somewhere, where they tried to forecast everything, waving their tails. I don't know where those tails went and where they fit in... That's where you belong. That's not the point here.
Why not all at once? Take the beveled t-distribution and go....
I don't get it. What do you mean by "Why not all at once?"?
If the tail, it's automatically traded, it doesn't need to be traded in any special way for that. And the centre, out of 50 pips of movement, we will take 30-40. Taking into account possible entry errors, we will get ~20 pips per trade, on average.
Or do you mean market exit? In general, I don't understand, clarify.