From theory to practice. Part 2 - page 101

 
denis.eremin:

No.

Aleksey Nikolayev:

What a joke, the usual martingale property (SB without drift is the simplest case of a martingale).

Purely abstractly yes, if the series is infinite and if you sum the increments, but the market series is closer to geometric wandering, and if you take even the usual Wiener process, there is still no practical sense in infinities, and the expectation of finite parts does not converge to anything, neither to zero nor to the initial values. Infinite series, infinite sums, all this must be forgotten, the sooner the better. We mere mortals, with finite life span and finite capital only care about predictive models of certain statistics, not generated by some abstract properties of series from some distributions, but generated mostly by economics and behavioral mechanisms, which are closer to alchemy than to rigorous theories.

 
J.B:

Purely abstractly yes, if the series is infinite and if you sum up the increments, but the market series is closer to geometric wandering, but if you take even a usual Wiener process with vanishing into minus, there is still no practical sense in infinities, and the expectation of finite parts does not converge to anything, neither to zero nor to initial values.

Well, glad you've finally discovered the law of large numbers.

P.S. If we take "finite plots", then the notions of MO and variance are worthless at all. Neither is the entire theorist.

 
denis.eremin:

Well, glad you've finally discovered the law of large numbers.

P.S. If you take "finite plots", the concepts of MO and variance are worthless altogether. Neither is the entire theorist.

Do not confuse large numbers (20-200 and above) when classical statistics makes any sense and infinite numbers that are interesting only to theorists and students yes session. SB MO on finite sections is not predictable, variance is partially predictable and much more stable.

 
J.B:

Purely abstractly yes, if the series is infinite and if we sum the increments, but the market series is closer to geometric wandering, but if we take even the usual Wiener process with drift, there is still no practical sense in infinities, and the MO of finite sections does not converge to anything, neither to zero nor to the initial values.

If I'm not mistaken, geometric SB (in the absence of drift) is also a martingale. Probably it's not the absence of expectation there, but the fact that its estimate does not converge to it due to an increase in variance.

 
denis.eremin:

Well, glad you've finally discovered the law of large numbers.

P.S. If you take "finite plots", the concepts of MO and variance are worthless altogether. Neither is the entire theorist.

I assure you, in order to calculate, for example, ACF SB no infinity is needed, but a basic knowledge of the theorem would not hurt at all)

 

I look in the book and I see... history.

SB is subject to laws, just as everything in the world is subject to laws.

In the world, in the universe, everything is alive, even machines, metal and plastic rockets are alive. After all, they can move, they just need fuel to power them.

Even unmanned iron cars move without a living organism, they move only with the help of a program and fuel or, roughly speaking, nourishment.

So also a human being if not fed will soon stop and lie down, stretching out his legs))

========

For the big players the fuel is the speculators.

If interested, I will continue later)))

 
Aleksey Nikolayev:

If I'm not mistaken, geometric SB (in the absence of drift) is also a martingale. Probably it's not the absence of expectation there, but the fact that its estimate does not converge to it due to an increase in variance.

Here and I do not remember, and to remember it is not much, as an experienced physician does not remember word for word the oath of Hippocrates, or Latin names of all organs from the anatomy course, there are of course unicums with super-memory, but the practical use of them as of Wikipedia, then either reflections or memory, more one less the other. The most important thing for SB is that it is impossible to predict the future increase of trend \return and that is it, it makes trading impossible, no matter whether it is trending or reversing, all other statistics anyway should indirectly predict the price change (when one moves away from the Mo = price turns around, etc.) and it is impossible due to the nature of the series. With volatility the situation may seem better but in fact it is the same, option trading on SB makes no sense because vol is constant, trading will stop.

 
J.B:

I do not remember, and I do not need to remember much, as an experienced doctor does not remember the Hippocratic Oath or the Latin names of all organs from anatomy course verbatim, there are of course unicums with super memory, but the practical use of them as from Wikipedia, there is either intelligence or memory, more one less the other. The most important thing for SB is that it is impossible to predict the future increase/return and that is all, it makes trading impossible, no matter whether it is trending or reversing, all other statistics anyway should indirectly predict the price change (when one moves away from the Mo = price reverses, etc.), but it is impossible due to the nature of the series. With volatility the situation may seem better but in fact it is the same, option trading on SB makes no sense, because the vol is constant, the trade will stop.

The impossibility of making money on SB is an obvious fact. There was a question here in the thread - what is the point for traders in this fact? For me (and for many others) it makes sense and is very simple - we need to look for price differences from SB. This approach is formalized by means of methods of hypothesis test theory from matstat. What I wrote earlier about zigzag is one possible example of this very standard approach. I have another example in my article about gaps.

 
Uladzimir Izerski:

I look in the book and I see... history.

SB is subject to laws, just as everything in the world is subject to laws.

In the world, in the universe, everything is alive, even machines, metal and plastic rockets are alive. After all, they can move, they just need fuel to power them.

Even unmanned iron vehicles move without a living organism, using only software and fuel or, roughly speaking, nourishment.

So also a human being if not fed will soon stop and lie down, stretching out his legs))

========

For the big players the fuel is the speculators.

If interested, I will continue later.))

Ararat 3-star cognac seems to have quite a foothold in the body too

 
Maxim Dmitrievsky:

Ararat 3-star Cognac seems to have taken root too

It's a bit expensive for a pensioner.

According to my observations, booze that can be identified in any way costs from a thousand roubles.

ZS: we have a whole range of so-called distillers at industrial shops (household, construction, electrical ...) now everywhere? I think that the distiller + a few pounds of sugar is practical, people take it actively, there are even specialized shops distillers, I think I saw or advertising or heard on the radio - the name of the shop "Pervachok" ))))

Reason: