Testing real-time forecasting systems - page 23
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
How about going through the history tester and gathering statistics?
I assure you, it will not be that simple, because the problem of accurate model identification will not disappear. Prediction of such indicators is not better and does not simplify the problem.
to lea
I don't know what you're predicting with, but for sure with some parametric model. Try to go through the history, adjusting your parameters to the best ones at each iteration of the forecast. And look at the "chaos" you are likely to see.
And you don't need anything else for absolute happiness...
I assure you, it will not be that simple, as the problem of accurate model identification will not go away. Predicting similar indicators is no better, nor does it simplify the problem.
to lea
I don't know what you are forecasting with, but you must be using some kind of parametric model. Try to go through the history, fitting the best parameters at each iteration of the forecast. And look at the "chaos" you are likely to see.
Depending on the two parameters you may get an outlier on the first bar towards the zero line (sometimes it breaks the forecast). On the whole, I don't see chaos.
P.S. I do not consider the idea of predicting indicators to be a serious one. I cannot make something nice, though I have already written a simple library for data analysis.
The first results of my prediction
Impressive.
to granit77
А для абсолютного счастья больше ничего и не надо...
Yes, but they are the most troublesome. Pivot levels are places where the price happens very rarely, there are few statistics and it is very "fuzzy". Such "black holes". The accurate prediction of such levels is a great art bordering on shamanism. And not accurately forecasting them is of no use.
to lea
Depending on the two parameters you can get an ejection on the first bar towards the zero line (sometimes collapses the forecast). In general, I do not observe any chaos.
I wasn't referring to the forecast itself but to the behaviour in time of the model parameters when the most optimal forecast was obtained (almost exact coincidence with the fact).
PS: By the way, prediction again by eye, not very correct, fact does not behave that way
Good morning everyone, found a new interesting index on the file factory, Pattern recognition is called
it looks like this, it's a pattern recognition predictor, i used to order one of these..... but i don't know what came out,
but i like this one a lot.
original text by the author:
I'm currently experimenting with a very simple pattern recognition. It takes n bars (or more exactly the value of the SMA, period 2 at each of these bars), calculates the difference between each of these values and the last bar in this group of n bars, then it normalizes this n-dimensional vector to length 1. It does this with all bars in the specified range, so it basically creates such a vector out of *every* bar with it's preceding n bars, so if n is 60 then it will make a vector out of bars 0..60, 1..61, 2..62 and so on.
It will then take the current pattern vector (the last n bars) and simply calculate the euclidean distance to every other vector (and to their inverse vectors to find mirror images) out of the history and plots the portion of the historical chart with the closest match in the indicator window.
Feel free to improve the code (I'm sure there could be some speed optimizations done), experiment with it, suggest better algorithms, earn big money with its "predictions" and buy me a beer once you have become rich ;-)
PS: The second indicator in the zip file will plot the second or third best match. It communicates with the first one via global variables, so the calculations must only be performed once.
у меня маленькая брозьбачка, сделайте плз. отображение первого инда (TotalRecall.mq4) в главном окне.
and, of course, give your thoughts and ideas (^__^)
the constant redrawing of the data...it means you can't really track the direction of the price, if the price goes up - you get a smooth transition down then up, if it goes down there's a deep dip down...
I now have my forecasts up to price swing level on H4 more reliable and robust, for a more accurate move I use M1 with exact pivot point from the channel boundary...'EURUSD in a couple of weeks from 1.4 to 1.5'
What about doing it in the main window, is that a hardship?