Forget random quotes - page 12
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
Let's distinguish the philosophical side of the issue from the analytical (algorithmic) specifics.
On the philosophical side, I argue that the price depends on sunspots. And let me be refuted, but in another thread.
In this thread I assert:
1. The price is being manipulated by dodgy citizens.
2. We work with time series, which we may or may not take into account time explicitly (e.g. periodicity), but all figures we work with are strictly ordered by time. Trying to take the topic away from sunspots is in another thread.
The supply and demand ratio (at a given point in time) is expressed by the price level (at that point in time).
What is in brackets must somehow be excluded from the equation, because it is an independent variable.
The presence of the time scale as an additional dimension allows you to trace the dependence between the supply/demand and the price in dynamics and to identify the properties of this dependence.
It is necessary to catch it, time is an auxiliary tool.
IMHO.
The supply and demand ratio (at a given point in time) is expressed by the price level (at that point in time).
So it's not a fact, it's a value that the price tends to through overshooting jumps, knowing this value, you can ignore the jumps, or use them to your advantage))
The correlation of supply and demand is hidden from us and masked by all sorts of filters.
So it is not a fact, it is the value to which the price tends through overshooting jumps, knowing this value, you can ignore the jumps, or use them to your advantage))
Naturally, this is a manifestation of inefficiency. There is a phase shift between supply/demand and price.
That's what we're trying to capitalise on).
As far as the filters will allow(.
Naturally, this is a manifestation of inefficiency. There is a phase shift between supply/demand and price.
That's what we're trying to capitalise on).
The understanding/justification of what? That time is not a prime mover? As if I myself don't know that when formulating equations of motion in, say, mechanics, real root causes are important - not time...
Yeah.
I've already suggested you initiate a new thread and tell others how you found the keys where you dropped them. Better with real-life examples - e.g. real-life monitoring.
I won't fall for that kind of persuasion. Money loves silence.
It was a simple illustration, don't be picky. You touched on an ancient scholastic problem. I illustrated that it was irrelevant.
I'm not convinced by your arguments, you're not convinced by mine, let's stick to ours.
How do you mean what? You yourself say this is not the time. You know better than that. And I also know that it's not the right time.
I don't know, and I don't givea shit about it. If you're a fundamentalist, you should say so. Sorry for the harshness.
What difference does it make now? I'm also looking for the root causes, among which there is no time. Have we reached a consensus?
OK.
Well, that's what I was talking about when I suggested rubbing out the flooders' posts.
Let them be. See, their monitors are already cracking - the root causes don't fit.
At least I was able to articulate more or less clearly the role of time in time series.
The issue of periodicity was little clear to me due to years of using TA. And only after I switched to analytics did I manage to approach it.
Once I detrended a quotient using a new method, but in addition to trend removal it automatically removes cyclicity. Then I was surprised to see that the parameters corresponding to the cyclic trends are not equal to zero! It was EURUSD H1 and no cyclic activity was visible, while the algorithm detected cyclic activity. At the same time the quality of the residue has improved.
That's why I paid so much attention to the time dependence in quotes.
In my several posts I refer to specific examples of models where time stands explicitly as an argument. Instead you are about caring about your monitor.
In statistical analysis of a quotient, the basic technique is to detrend the quotient, since the presence of a deterministic component distorts any statistics. The standard technique is to include a trend in the model in the form of a*t or a*t2. This is usually enough. Here time is directly as an argument of the function.
You cannot strictly prove the existence of a*t or a*t^2 trend. When estimating model parameters with such terms at different data points, the coefficients will differ statistically significantly (in previous discussions I suggested checking this with Chebyshev inequality), i.e. there is not estimation error, but incorrect model specification - in this case, even R^2=0.99 tells us nothing.
I see, you are a fundamentalist.
Can I not answer that question? Definitely not the time itself.