Finding local highs and lows in price - page 7
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
Yes, there are as many ways to consider the U-turn as you like (what to build ZZ on). It's been written about a hundred times already. No, more...))
There is indeed a lot of writing. But I started with something else. Two types of forecasts are known: of price and of price direction. I, on the other hand, offered to discuss the forecast of ZZ0 reversal, i.e. the forecast of the past estimate. I don't recall anything like that. What we discussed above is a pattern, but it does not give a prediction probability or confidence interval.
Yeah. I don't recall that either. But it sounds - roof-breaking: predicting an assessment of the past.
AH... UH... I don't know what to say. I must have misunderstood something. Otherwise, I need a drink. To get my head and this "other" together...)))
===
Estimation of the confidence interval (both in price and time) can be given by the points of crossover of Fibo (or D) levels with the Fibo channel. And quite accurately. Well, they are of course built from extrema that have been formed. One can draw a fan of such null rays from 1 (ready) extremum to those points.
An estimate of the confidence interval (both in terms of price and time) can be given by the points at which the Fibo (or D) levels cross the Fibo channel. And quite accurately. Well, they are of course built from extrema that have been formed. It is possible to draw a fan of such null rays from 1 (prepared) extremum to these points.
If we believe in Fibo. A market model such as the ARPSS is needed. But it predicts the future as it should. Perhaps this model should be used to estimate the probability of extremum by ZZ? In theory the forecast accuracy should increase with each candlestick(confidence interval becomes narrower).
Of course there have been such studies, I know that :). It's another matter that they haven't been much discussed. The confidence intervals there are usually such that the actual prediction does not make much sense. Although maybe I just took the wrong parameters :)
If we're talking about ARPSS, this model has been used in econometrics for about 30 years, but for some reason it has never been discussed on this forum. Similar issues were discussed, but not as systematically as it was in ARPSS. I cannot understand why.
If we talk about ARPSS, this model has been used in econometrics for about 30 years, but for some reason has never been discussed in this forum. Similar issues have been discussed, but not as systematically as ARPSS. I can't understand why.
I think it's because they're looking for grails here. In 30 years it has become clear that there is no grail there :)
As far as I am concerned, ARPSS is not a grail, a time consuming and not all sites applicable model, requiring skill and experience in application. But it is better than idle speculation about some parts of this model, an attempt to solve problems that have already been solved in this model.
I don't know. To me, this sort of thing seems to make no practical sense. Although, there may be a lot I don't know that I should have known. The thing is, apparently, that I make my living from trading and am therefore extremely pragmatic in my approach.
Perhaps it is worth revisiting. Sometimes I have left a topic that seemed uninteresting at the time and then, thanks to my new knowledge, it no longer seems so unpromising.
But what I arrived at in trading was not the first thing I came across either. Many things have been tried. Perhaps something has been studied superficially.
It is never too late to learn. The main thing is to have something to live on. Nobody will pay my scholarship but myself...))
===
And the fact that there is no grail was discovered long before the blood of the Nazarene was collected there. )))
One can think and search endlessly, and the search itself becomes a goal. A fragment of Hamlet's soliloquy (translated by M. Lozinski):
"...So cowardly does our reflection make us,
"And so the colour of nature decides
"So cowardly is the colour of our natural resolve,
And beginnings, so mightily stirred up,
"andturn aside their course,
They lose the name of action..."
I'm just talking about the whole thing. You can come across a lot of different things here on this forum.
))) Somebody sincerely thinks that the aim of trading is to keep a balance; somebody thinks the aim is to find a way to trade; somebody likes more abstruse constructs... etc.
And the goal of trading is the same - stable profit. And in reality it's easier than many people think. Although, maybe it's not so interesting, not very romantic and somehow clumsy.))
===
Petyka's got a lot on his plate... )))))))))))))))
It's never too late to learn. The main thing is to have something to live on. Nobody will pay my scholarship but myself...)))
+5.
I used to know an acquaintance of mine, whom I met at the age of 19 - he started in the market. He knew how to write indicators, learned to write TS while I was there, but he never used it - he didn't use anything at all. He had a depo = $50k. When he needed money (for his limited needs), he would turn on the computer, maybe wait, get into a position and then get out. Skimmed his winnings. Almost never lost. But that's the only one I know.
In my TS system I naturally use some indicators and algorithms, but the range of tools I use is much wider and they are used for better understanding of the market. Different tools allow to see more clearly different sides of the market.
I think that ARPSS will advance my understanding of the market. Will it be possible to use it directly to make a profit? - I think you can, but judging by the publications, not always. But also my TS does not always make a profit and I cannot understand why it makes a profit in some areas and not in others. I have to rely on some indirect evidence of functionality, like profit factor. But there is no understanding of the essence of the problem. The ARPSS will know exactly why the model is not working.
There is a parallel thread discussing the predictions (the thread is called Optimize Testing! https://www.mql5.com/ru/forum/115498). Forecasts are published that have nothing to do with the subsequent movement. And there's no questions like: "Is it possible to make forecasts at the given market point? Or about the confidence interval?
I myself liked my idea to predict (calculate the probability of) a ZZ reversal, because ZZ is already in the past, and for such a prediction we have information future in relation to ZZ and past in relation to the prediction point. If the algorithm did not give a prediction, then we should calculate on the next candle, etc.