Machine learning in trading: theory, models, practice and algo-trading - page 342
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 typical mistake of all radio engineers is that they think that there is a signal on financial markets, but they can't imagine even in their wildest dreams that there is no signal on financial markets and there never will be. This is why filters are almost never used in financial markets.
There is another, purely technical circumstance: financial markets are non-stationary time series, and as a result, the lion's share of statistics, which work fine in radio engineering, goes into tatters. I have outlived so many radio engineers on this forum. I advised them all: if you want to make money, forget radio engineering. Forever.
Typical reaction that comes from ignorance of the subject...
Typical reaction that comes from ignorance of the subject ...
Why have you turned off all monitors? where to get wisdom now)
What kind of patterns are you looking for in radio engineering?
In radio engineering, the existence of a regular sequence, such as a TV picture or music .... reflected locator signal. The signal may be scrambled. But it is the existence of the signal is postulated - it is known for certain.
We should add to the differences that in radio engineering we recognize the past, and in trading we are only interested in the future, and the past is only at the stage of development.
The mainstream among amateurs in financial markets is the search for what is not known. The idea of patterns follows the postulate of technical analysis "history repeats itself". On the level of belief, we are looking for some kind of fuzzy patterns that change over time. At the same time we believe that the pattern will unambiguously correspond to the trade order. All of this at the level of belief is a heavy legacy of technical analysis.
Within the application of mathematics to trading, there are actually two directions:
1. Classification. The target variable coincides with the orders in the terminal.
2. Consecutive modeling of the input quotient: smoothing (filtering); looking at the whole residue, modeling this residue; looking at the residue ..... Stop if the residual is stationary. If successful, then the prediction works a few steps forward. If not, we again shamanize on non-stationary markets and drain the deposit.
A fundamentally different approach. Incorrect choice of tools (matlab, matcad, ... filters and different Fourier tools) aggravates the problem.
So, once again.
Forget radio engineering.
So, once again.
Forget radio engineering.
I'm not aware of your practical results, but my radio engineering approaches showed their consistency a relatively long time ago.
The real result of the application of TS in the market - the probability of a successful transaction is not less than 0.5, with a profit/loss ratio of not less than 2/1.
So, let's forget about radio engineering).
Why did you turn off all the monitors? Where to get wisdom now)
Well, you found yourself a guru ;)
Typical reaction that comes from ignorance of the subject...
I am not aware of your practical results, but my radio engineering approaches showed their consistency a relatively long time ago.
The real result of the application of TS in the market - the probability of a successful transaction is not less than 0.5, with the profit/loss ratio not less than 2/1.
So, let's forget about radio engineering).
Success, I wrote from the bottom of my heart.
There is such a university specialty "Automated (as opposed to automated) control systems". Instead of forum, you sit on the bench, learn the basics, expand your horizons, learning that in addition to random and deterministic processes there are indeterminate processes. You get a phenomenal result in your old age: you stop embarrassing yourself on the forums.
You're being silly again...
And you did not get a phenomenal result in your old age: you do not stop embarrassing yourself on the forums.
Success, I wrote from the bottom of my heart.
Thank you. Of course, and in your research success. I do not consider my approaches to be the only correct one, and I admit that there are other, possibly more effective approaches to the tool.
However, there is no doubt that Data Mining, at some point becomes an absolutely necessary element of the system. However, the choice of specific methods is a problem.)
I pinned my hopes on the NS, but it does not seem to work. At least a simple experiment with real data failed. Let's look for... with mother-of-pearl buttons.)
ZZY It is interesting that in this case, the NS was perfectly trained and recognized (classified) the artificially created patterns.
You're being silly again...
And you didn't get a phenomenal result in your old age: you don't stop embarrassing yourself on the forums.
I haven't seen a single decent result here yet.
Probably wondering - it's new...
But no one and nothing worthwhile has been found yet.