Machine learning in trading: theory, models, practice and algo-trading - page 2529
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
It seems to me it is impossible to wrap numerical optimization in a formula.
No well probably you can apply stochastic integrals or something like that, but it's too hardcore if you can just opt out.
It is clear that the counter-trend. What is the specific algorithm?)
Is there any particular algorithm for H > 0.5?)
It is obvious that the instrument more inclined to the return should be traded on the return, the only difficulty is that this indicator floats (quasi phase) and the deviation from 0.5 is sometimes not strong enough, and the calculation of H itself has a window effect, as a result anyway some kind of additional analysis is needed.
Probably the practitioners won't write here their results, but among the obvious ones is for example trading in night flat, if the broker is not too greedy with spread and there is no significant news in Asia-Australia that night.
Everyone knows about night scalpers, but that's how many times I didn't count Hearst=0.5, just because of lesser number of ticks the vol goes down.
We know about complexities) To simplify, let's set complexities equal to zero. What is the formula to trade the return for maximum profit and minimum drawdown?
In ideal conditions (flat) we draw the distribution and calculate the level with the maximum profit. It is more difficult to calculate during averaging.
It seems to me it is impossible to wrap numerical optimization in a formula.
Is there any particular algorithm for H > 0.5?)
In general, you need to have a deviation greater than X%/points
Well, if you can derive the ACF SB formula, I think you can derive the equities formula for similar processes as well.
Maybe nobody solves it 🤔 just to calculate the equity of trade you should either simulate this trade or make some empirical model of equity dependence on parameters...
Deviation from what?
From a certain starting point or from the average or from something else, there are many variants and all of them are probably equivalent, in a word, to register a price movement of some percent or points.
There are a lot of algorithms, the simplest one is: "went up - buy".
Well, with this interpretation of "specific algorithm" here is a specific algorithm: with H < 0.5 "fall - buy" )).
I wonder how it should be by science.
For example, a channel break.
From some starting point or average or something else, there are many variants and all of them are probably equal, in a word, to register the price movement in so many percent or points.