Machine learning in trading: theory, models, practice and algo-trading - page 2519
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
I do not work on this scale, alas. I have a window of 8 hours and above astronomical time (and, accordingly, variable market time)
Replace minutes with hours and it will be yours. the question remains.
If I wanted to ask what time interval should be used to measure the speed of ticks. Bigger, smaller, equal to TF?There's a type of prediction in the market called "overlapping the gap"
The target is time-independent, something like - "maybe now, maybe after tomorrow, but the price will be above/below this current point with 100% probability".
What if I look for patterns that can predict with 100% probability such an event, you don't know if the price will go through now or in two days, but if you have a forecast with 100% probability and if the price hasn't broken through now, it could be a good guide for trading tomorrow or the day after that...
There's a type of prediction in the market called "overlapping the gap"
The target is time-independent, something like - "maybe now, maybe after tomorrow, but the price will be above/below this current point with 100% probability".
What if I look for patterns that can predict with 100% probability such events, we don't know if the price will go through now or in two days, but if there is a forecast with 100% probability and if the price hasn't broken through now, it could be a good guide for trading tomorrow or the day after...
Target, we don't know when in a certain period of time it will happen. And if we don't know when, and there is no time limit, then it's a bad target, even if it's 100%.
A target, we don't know when in a certain period of time takes place. And if we don't know when, and the time is not limited, then it's a bad target, even if it's 100%.
You can limit it to a few days, a week...
The point is not to predict what will happen on the next candle (it's impossible to reach a true probability there, even 70% is already out of space), but to predict the event with a 100% probability, even if it won't happen right away... And then build a model of price behavior on it
There's a type of prediction in the market called "overlapping the gap"
The target is time-independent, something like - "maybe now, maybe after tomorrow, but the price will be above/below this current point with 100% probability".
What if I look for patterns that can predict with 100% probability such an event, yes we don't know if the price will go through now or in two days, but if there is a forecast with 100% probability and if the price hasn't broken through now, it could be a good guide for trading tomorrow or the day after...
We can live with the uncertainty of time. The problem is always with the uncertainty of how long the price will go in the opposite direction before it reaches the forecasted point.
For example, in the case of SB, any point will be reached with unit probability, but it does not give an opportunity to earn.
You can deal with the time uncertainty. The problem is always with the uncertainty of how much the price will travel in the opposite direction before it reaches the forecast point.
For example, in the case of SB, any point will be reached with a unit probability, but that does not give an opportunity to make money.
I agree, but knowing that the price will be at a certain price with 100% probability within a week, it's better than not knowing anything, right? I have no idea what to expect, where to take a strike.
I agree, but knowing that the price within a week will be at a certain price with 100% probability, it's better than not knowing anything, right? At least we know what to expect, where to take the strike.
The time limit will decrease the probability.
There's a type of prediction in the market called "overlapping the gap"
The target is time-independent, something like - "maybe now, maybe after tomorrow, but the price will be above/below this current point with 100% probability".
What if I look for patterns that can predict with 100% probability such events, yes we don't know if the price will go through now or in two days, but if there is a forecast with 100% probability and if the price hasn't broken through now, it could be a good guide for trading tomorrow or the day after...
I agree, but knowing that the price within a week will be at a certain price with 100% probability, it's better than not knowing anything, right? At least we know what to expect, where to take the strike.
Yes, half of the problem is solved, the other half is where to take a stop loss.