Machine learning in trading: theory, models, practice and algo-trading - page 425
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, you are right absolutely. Does anyone really teach the model on a zigzag? What's the point of teaching the system to predict any indicator (even if it is not a peeping one), if we trade the price and not the indicator.
We trade the price, and the decision comes on the basis of some situation indicated by the indicators (I apologize for the tautology, I meant it).
This target absolutely depends on the quotes as well as the independent one. Yes, the quotes with different time stamps. But it is also calculated on the basis of quotes from the future. I do not understand your logic and objections on ZZ.
who use ZZ and similar indicators as a target and are happy with 70-90% accuracy of predictions.
Don't be so harsh and distorting.
It would be much more useful for you to understand the meaning of what I've written.
The solution comes from our interpretation of this indicator's readings. So, first you teach the system to predict some indicator, and then you will have to teach how to interpret its readings correctly in order to make profit.) Oiled up. Why the extra link?
The solution comes from our interpretation of this indicator's readings. So first you teach the system to predict some indicator, and then you should teach how to interpret this indicator correctly in order to make profit.) Oiled up. Why the extra link?
You're right, predicting the indicator is an unnecessary link. But predicting the price is also an unnecessary link, because the goal of the trading strategy is profit, not price.
Following this logic, we should model trader's behavior, not indicators or price charts, in this case the best target function is profit trade results.
And practically, we take the trading history or run the advisor in the tester and teach the model - price patterns on the input and trades on the output of the trading report.
You're right, predicting the indicator is an unnecessary link. But predicting the price is also an unnecessary link, because the goal of the trading strategy is profit, not price.
Following this logic, we should model not indicators or price charts, but the trader's behavior, and the best target function would be the indicators of his profitable trading.
In practice we take the trading history or run the Expert Advisor in the tester and train the model - price patterns on entry and trades on exit.
That is really the most important thing, there is no reason to build a separate forecasting model, it makes much more sense to build a profitable model, which consists not only of forecasting, but also of other modules that should also be intertwined with forecasts. And what is being discussed here this is some kind of kindergarten, use zigzag or not, what the hell difference does it make if the approach with forecasting is initially wrong (i.e. does not lead to success) :)
Somebody somewhere advised to put "IMHO" everywhere.
So, this is IMHO.
This is really the most important thing, there is no point in building a separate forecasting model, it makes much more sense to build an earning model which consists not only of forecasting but also other modules that should also be somehow intertwined with forecasts... And what is being discussed here this is some kind of kindergarten, use zigzag or not, what the hell difference does it make if the approach with forecasting is initially wrong (i.e. does not lead to success) :)
Someone somewhere advised to put IMHO
this is my opinion
If this is banter, as is customary here, I will add that behind the words there is also some experience in this direction.
Where's the banter in there? This one is also from personal optics
As Father Mandelbrot still wrote, - predicting the price is the way to collapse, but you can estimate the likelihood of future volatility.
You don't have to twist things around so crudely.
I dare not do so, not in your attitude, nor in anyone else's.
It would be much more useful for you to understand the meaning of what I have written.
There is a lot of sense in what you have written, the article is great, I mean it absolutely sincerely, it is concise and succinct, respect! But there is a little flaw, it is what I pointed out, as I myself have been deluded in the same way for a long time.