The riddle: the distribution bell rattles - the broker says the price, whoever it hits sheds tears and loses their deposit. - page 4
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
Exactly!) does not make sense until it is not clear what is being analysed)
(Randomness also has regularities.)then randomness is not random
;)
Exactly!) It does not make sense until it is not clear what exactly we are analysing)
(Randomness has patterns too)What exactly and more importantly for what exactly!
What is the purpose of this analysis?
Profitable trading, no?
Exactly!) does not make sense until it is not clear what is being analysed)
(Randomness has regularities too.)Hopefully, methods of probability theory (based on Kolmogorov's axiomatics) will be used to analyse patterns of randomness. Other methods tend to amount to demagogy.
Hopefully, methods of probability theory (based on Kolmogorov's axiomatics) will be used to analyse patterns of randomness. Other methods tend to be reduced to demagogy.
Exactly.
Exactly!) does not make sense until it is not clear what exactly we are analysing)
(Randomness has patterns too.)Let's deal with this randomness. In a random process, the outcome of luck/failure is distributed roughly 50/50, and in our case, this is far from being the case. Consequently, the market is not random but chaotic, so the search for patterns does not give the desired result. On the other hand, traders' appetites are much higher than an exceptionally perfect market can afford. Nevertheless, of course there are certain patterns. The trouble is that no one wants to go into the essence of regularities found by someone. Certainly, everyone wants to find his or her own pattern and that dissipates the forces and opportunities of market researchers. I recently voiced 3 areas of pattern-finding - zero attention from forum participants. There is no real discussion, everything revolves around finding patterns of random processes at tick and minute level. I'm sure the patterns show up over much longer periods of time.
With a random process, the outcome of luck/failure is about a 50/50 split, but in our case, this is far from the case.
It doesn't have to be. Imagine a "shifted centre of gravity" coin that has a higher probability of one side falling out, for example 2/3, and the other then: 1/3. Suppose the falling out of the first side corresponds to an addition and the second to a subtraction of 1. Then we get:
1) an upward trend
2) the independence of the incomes (there is a misconception that the existence of a trend necessarily implies the dependence of price incomes)
3) we will see a "bell" of incremental density distribution for N passes (shifted to the right comparing to the symmetrical coin, it seems to be written about binomial distributions)
Now imagine that the offset of the centre of the coin may sometimes change (by whoever throws it), then:
1) trends will change each other unpredictably (for spectators)
2) the independence of increments will be preserved
3) the distribution bell will remain, but with an unpredictable shift (and not binomial).
It will be a process with independent non-stationary increments.
Not necessarily. Imagine a coin "with an offset centre of gravity" that has a higher probability of one side falling out, e.g. 2/3, and the other then: 1/3. Suppose the falling out of the first side corresponds to an addition and the second to a subtraction of 1. Then we get:
1) an upward trend
2) the independence of the incomes (there is a misconception that the existence of a trend necessarily implies the dependence of price incomes)
3) we will see a "bell" of incremental density distribution for N passes(shifted to the right comparing to the symmetrical coin, it seems to be written about binomial distributions)
Now imagine that the offset of the centre of the coin may sometimes change (by whoever throws it), then:
1) trends will change each other unpredictably (for spectators)
2) the independence of increments will be preserved
3) the distribution bell will remain, but with an unpredictable shift (and not binomial).
It will be a process with independent non-stationary increments.
Here I agree.
3) there will be a "bell" of the density distribution of increments for N throws (shifted to the right compared to the symmetric coin, about the binomial distribution it seems to have already been written here)
To my observations "behind the bells", the price always comes back to close the nearest historical minimum/maximum intraday - there are rumours that they are hunting for stops, but it's much easier, it's the job of market makers - they have to buy back everything and sell everything, that's all I can say in brief:
https://www.mql5.com/ru/forum/119024#comment_3151140
After all the current sell/buy offers are done, there will be a jump to a new level, and the price will go back to closing new highs/minimums.
And in your model with a coin - in the long term it may work, but it's not all smooth sailing - there are expiration dates, options, interest rates...and the market has to take it all into account and act accordingly.
so it turns out that there are a lot of market participants and the price moves considering everything
ZZZ: Your model with a coin is likely to work on a weekly timeframe - just as price went 3-5 cents up and came back 3-5 cents down, that is the scale of 1 cent, rather than looking for intraday price changes of 0.01 cents
There will be a bell, but no one knows where it will be shifted to (or rather the shift of the bell means nothing), too many market participants, according to my observations "behind the bells", the price always comes back to close the nearest historical minimum/maximums within a day - there are rumours that they are hunting for stops, but it is much easier, it is the market makers job - they must buy back everything and sell everything, so it is clear and brief:
https://www.mql5.com/ru/forum/119024#comment_3151140
After all the current sell/buy offers are done, there will be a jump to a new level, and the price will start to move again to close new highs/minimums.
And in your model with a coin - it may work in the long term, but it's not all smooth sailing - there are expiration dates, options, interest rates...and the market has to take it all into account and act accordingly.
so it turns out that there are a lot of market participants, and the price takes into account everything
That bit that you have quoted is in fact a trader's paradise (trend is your friend). Therefore it is unlikely to be pronounced and lasting) The second part of my post seems to me a bit closer to reality.
It is believed that a certain symmetrical bell is formed, but in the asymptotic sense(there is no special practical sense in it)
The essence of any model is to simplify a phenomenon in order to consider aspects of it that interest us. For example, if we start to build a theory for options, we will also need simplifications, but they will be different.