The market is a controlled dynamic system. - page 34

New comment
 
Tantrik:

The candlesticks are smaller, the monitor is larger (I've seen the charts myself on widescreen monitors that are 1.5 metres long), and the bars are displayed in the settings. (Why do you need three years for H1?)
To calculate the slope on three years. See above.
 
Tantrik:

reduce candlesticks, monitor larger (I've seen the charts myself on widescreen monitors that are 1.5m long), display bars in the settings.(Why do you need three years for H1?)
so the monitor does not get idle ;)
 

1. Put the hubris to rest. You are not market makers. You CANNOT influence the market. Suck it up or go fuck yourselves.

2. It follows from point 1. All you can and what you should do if you want to make money, is to learn to recognize what the sinister financial behind the scenes is doing FOR you. Then maybe you can get something back from this "hostile" reality. Which is not a hostile reality - what would you do if you weren't given a reference point? )

3. Knowing how to trade is, for the most part, a philosophy. How to build an MTS - the answer lies in your attitude towards the market. points 1,2.

4. Here's what I don't understand at all, is trying statistical analysis of penguins, OTO, vanilla ice cream and what I did last summer. All rolled into one.
You know the expression, clever fool? What are you analysing?

 

I've been trying to make such a simple point... So far to no avail...

I'll rephrase it as a problem ;)

.

Suppose there is an object with transfer function W.

.

For illustration, we can imagine that this object is an oscilloscope.

Suppose that the oscilloscope inputs some unknown signal X. We observe the oscilloscope output signal Y on the display.

.

Let us further assume that we know the transfer function of our oscilloscope


The problem is as follows: We have to determine the input signal X.

 
avtomat:

I've been trying to make such a simple point... So far to no avail...

I'll rephrase it as a problem ;)

.

Suppose there is an object with transfer function W.

.

For illustration, we can imagine that this object is an oscilloscope.

Suppose that the oscilloscope inputs some unknown signal X. We observe the oscilloscope output signal Y on the display.

.

Let us further assume that we know the transfer function of our oscilloscope


The problem is as follows: We have to determine the signal X.

It's pretty simple: take the inverse Laplace transform X=Y/W. Why do we need to know the signal X? It is an external perturbation whose direction cannot be predicted. And knowing it in the past will not give us anything. Besides the transfer function W changes with time. Let's simplify the problem. The external perturbation is the momentum x(t)=u(t). First we must learn to identify the external perturbation at the output y(t) (i.e. by the movement of the price). It is quite simple: the rapid price change at predetermined times of news releases shows the presence of an external disturbance. Between perturbations (news) players react to this news. Our task is to know the direction of the perturbation and the initial reaction plot, and then predict the rest of the reaction before the next perturbation. The nature of the reaction is inconsistent, which makes our task difficult (otherwise everyone would be rich).
 

I have purposely simplified the task to the limit - so that the idea is clear.

Why do we need to know the signal X? -- Because it's the current reference signal, otherwise known as the control signal. You can, of course, think of it as an external perturbation. But it can also be seen as an element of the X-W-Y system. (we do not touch the prediction issues yet) In such an interpretation it will be enough to know the position of input X with respect to output Y in order to determine the further movement of output Y.

Further, PF W does change in time (but these are slow parameters - that's what I mentioned a couple of pages ago) -- here's the next problem: how to do without explicitly specifying PF -- but that will come later, at this stage it is enough to understand in general the principle possibility of such a description.

Introduction of pulses does not simplify, but on the contrary, considerably complicates the problem, because at once it transfers it from a class of linear problems to a class of pulse ones --- at this stage it is superfluous and does not justify itself - when there will be a model, then it will be easier to transfer it, if it will be necessary.

 

And for clarity's sake

 

and in continuation, a further development of the situation

 
Next, what is left to do is solve the transport problem?
 
- I don't care how you go. But you have to deliver all the presents and be at the appointed place at the appointed time.
1...272829303132333435363738394041...551
New comment