Creating a positive IO - page 7

 
yosuf:
Consequently, we should try to find a function capable of describing, even for the time being, the history of quotes, with an acceptable accuracy for practical purposes. I would like to know the opinion of the participants on the idea of dedicating a special thread to this issue.
As stated here earlier, if you find such a function, it will most likely become known to all. And since it will become known to everyone, its presence will soon be NEEDED by the actions of traders. Why look for it then...?
 
LeoV:

I agree that it can be done that way. But that might not work someday.....))))

someday the earth may fall into a black hole )
 
LeoV:

By studying neural networks and optimising/training various FCs (with and without neural networks), you should know how a positive MO in the past easily turns into a negative one in the future...))))

Fitting to the story, and using neural networks, you can fit 100% of the story to the tops - and trade on that story)) and it is unlikely to work in the future.

This is about something else. Suppose, looking at the quotes for one week in 2009 I've got a great idea. I programmed the algorithm without any optimizations and neural networks - and got the balance curve at 45C on the forward side up to this day. It's not a fitting after all.

 
  • examples of systems with positive expected payoffs;

1) In the foreign exchange market, a carry trade. On the stock market - "Buy & hold".

2) Arbitrage

3) Insider:)

In other cases, positive expected payoff is not guaranteed.

 
yosuf:
Consequently, we should try to find a function capable of describing, even for the time being, the history of quotes, with an acceptable accuracy for practical purposes. I would like to learn the opinion of the participants and devote a special thread to this issue. Moreover, the properties of this function should clarify many questions that have arisen here and now, and which have excited the participants due to their fundamentality.
Yusuf, so you seem to have found it. (18) is called. Or am I mistaken?
 
Alligator:

For example, like this.
My system in the past has a moderately high MO and let's say the number of continuous losses is 5.
I leave all parameters the same, but based on the calculation that my system will withstand without much trouble in the future
9-10 losing trades in a row. I am doing something like a safety margin. Plus purely from practice I don't see that even
that even this option (with 9 losses in a row) is possible.

The size of a series of losses can be calculated, but there is no exact algorithm that takes into account all the parameters. Basically, the size depends on TR and SL. The smaller it is, the longer is the series.
But the question arises: what will follow this series? Another series, smaller in size?
 
jelizavettka:
Yusuf, so you seem to have found it. (18) is called. Or am I mistaken?
No, you are not, but I have decided to show and prove its potentialities with simple examples, and when the participants raise their hands for this function, only then will I fully disclose its appearance. so that the opponents can be convinced of its simplicity, elegance and inevitability of success in the pursuit of price.
 
I apologise, of course, but you are trying, theoretically, to take profit away from all market participants.
 
DmitriyN:
The size of a series of losses can be calculated, but there is no exact algorithm that takes all parameters into account. Basically, the size depends on TR and SL. The smaller is it, the longer is the series.
But the question arises: what will follow this series? Another series, smaller in size?


Again, there could be lots of options.

For example, if there is such a series ( 10 losses in a row - although I hope you understand the probability of that),

I get a total loss of 20-30% on the account (pick it up) and decide to close it.

For the remaining 70-80% I open a new account.

 

and I decided to determine the trend by the mood of the hedgers, whose positions are characterised by counter-trending.

these big guys know better than we do what the trend is)