Is any prediction doomed? - page 33
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
Those fables were invented by academics who have never traded. Can we believe them?
My imho is that this is true, but the catch is that new factors can arise in addition to the existing old ones, and the speed at which they are accounted for at the moment can be different. It may be one hour candlestick of 300 points, or ten one-hour candlesticks of 30. That is, though the old is considered, but the new is not considered in the present, because it is only happening.
A forecast is a quantitative notion, operating with a fixed forecasting error that gives its limits, whereas a prediction is a qualitative notion, i.e. with an error of any value.
So many words, but no one wants to make a table for different prediction methods. To make five or six Expert Advisors for the most popular methods, put them on accounts and check if forecasting is doomed or not. There is no end to the chatter.
My imho is that this is true, but the catch is that new factors can arise in addition to the existing old ones, and the speed at which they are taken into account at the moment can be different. It may be one hour candlestick of 300 points, or ten one-hour candlesticks of 30. That is, though the old is considered, but the new is not considered in the present, because it is only happening.
A forecast is a quantitative notion which has a fixed error of forecast which defines its framework, whereas a prediction is a qualitative notion, i.e. it has an error of any value.
So many words, but no one wants to make a table for different prediction methods. To make five or six Expert Advisors for the most popular methods, put them on accounts and see if prediction is doomed or not. There is no end to the chatter.
I drew my conclusions on the basis of:
1. Spectral algorithms
2. Probabilistic algorithms
3. Perspectron algorithms (quasi-Newtonian learning)
-Aleksey-, list these "most common methods" here.
Regression;
Autoregression;
Harmonic analysis;
Autoregression of the integrated moving average;
Singular spectral analysis;
-Aleksey-:
...
A prediction is a quantitative concept, operating with a fixed prediction error that gives its framework, while a prediction is a qualitative concept, i.e. with an error of any magnitude.
....
Regression;
Autoregression;
Harmonic analysis;
Autoregression of the integrated moving average;
Singular spectral analysis;
:) so go ahead.
So I'm not an arguer, I'm a sympathiser.
That's it... screw it... prediction is doomed.
It's not. Forecast and prediction are the same thing. "Forecast" is a Greek root, it sounds more scientific.
Quite possibly, thank you, it's just that "prediction" is in the Math Dictionary, where it says error as part of it, but "prediction" is not, but it is in the non-technical dictionary, where there is nothing about error.
Prediction is a quantitative concept, operating with a fixed prediction error that gives its framework, while prediction is a qualitative concept, i.e. with an error of any magnitude.
Since the hedgehog has launched that word "prediction" in the current discussion, let's go back to his water tank. What kind of "error of any magnitude" does he have there, when he almost certainly knows what will follow. At the same time, operating with a fixed prediction error, he cannot predict.
So many words, but no one wants to make a table for different prediction methods. To make five or six Expert Advisors for the most popular methods, put them on accounts and see if prediction is doomed or not. There is no end to the chatter.