Optimal strategy under statistical uncertainty - unsteady markets - page 7
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
п.3.
If the outcome is the same as the one selected for the series, the series is over, go to step 1.
Otherwise i++, go to step 2.
---
And no history.
i++ is history of deals, since i equals e.g. 5, it means that there were 5 unsuccessful deals on eagle, and this is already statistics, which is forbidden in the problem condition.
The statistics will be simple as can be, and the statistics will be obscenely simple. Each time we bet on heads, if there is some profit it means all is cool - we continue in the same spirit, if the amount of money in the pocket began to decrease, it is necessary to "change the strategy" and always bet on tails.
We have already found out here that this is not true. The strategy exists and no statistics are needed.
We've already established here that this is not correct. The strategy exists and no statistics are needed.
Too bad I missed it. Reread the whole trid again, still missed it. Please tell me where.
We have already found out here that this is not true. The strategy exists, and no statistics are needed.
and the last dropout is not a statistic from a single observation? especially since this statistic is collected many times. you can collect a statistic once from 100 data or 100 times from one - it is still collecting and using statistics. so this example implicitly uses statistics from the whole row
Too bad I missed it. Reread the whole trid again, still missed it. Please tell me where.
On the first page, sixth post from the top.
Guys, I'm just at a loss as to what and how to answer this. The condition of the problem is given. The basic mathematical aspects are considered, in two views. It is stated that the results of the simulations confirm their correctness. It is said that this effect was observed on testing of real TS. Which aspect of the problem are we talking about?
I can give you some advice - don't respond in any way. There are several trolls on this forum, give them any proof, even links and quotations to any reliable sources, they will still violently defend only their point of view.
That is, if someone is talking outright rubbish, or after a reasonable response continues to stick to their beliefs only, then he must simply ignore and do not communicate. The purpose of a troll is to publicly defend only a personal point of view, no matter how much it contradicts reality. Ignoring a troll makes it clear that he has no audience - his blather is wasted because no one is paying attention and he moves on to other threads to find people to talk to and blather to.
It is said that this effect was observed on testing a real TS. What aspect of the problem are we talking about?
an example where this strategy worked ;) and in general, how do the conditions of this problem relate to the real market? :)
With information only about the coin, the possibility of making a profitable system will be as high as the known depth of history of the flip, it is from this history you can extract information not only about the coin, but also about the object performing the flip, and having information about the object, you can already guess not only about the side of the fall, but also about the time of fall, the spinning speed, etc.. I.e. it is possible with infinitely accurate probability to determine the next event of the coin, of course, as long as it depends on some object (which will ALWAYS happen no matter how tightly).
But if you know beforehand some number of factors (weight, material of coin, etc., etc.), then probability is the same, but time for calculation of big factors will change... If you look only at the previous event, you can hardly get far) So, you need to look for strategies that rely on deep history, especially if we only have a coin with an offset centre of gravity.
Since we will be betting on the previous outcome, i.e. only on the side that fell in the previous coin flip, respectively, p part of the bets will be on heads and q part on tails.
Since the probability of winning at an eagle bet is p, and eagle bets are only a p-second of all bets, the winnings from eagle bets are p * p = p^2Since the probability of winning for betting on heads is q, and bets on tails are only q - part of all bets, then the winnings of bets on tails are q * q = q^2
You're messing up something here, Jura. The winnings for equal bets (say 1) are simply p and q, but not p^2 and q^2.