From theory to practice. Part 2 - page 80

 
Alexander_K2:

Because the market series is more complicated than SB due to its non-stationarity, i.e. it is virtually impossible to build a stable earning TS on it, unlike SB.

So why all this talk about SB if you have to trade in the market anyway?

 
Alexander_K2:

Because market series are more complex than SB because they are non-stationary, i.e. it is almost impossible to build a stable earning TS on them as opposed to SB.

SB is a non-stationary process, just like price series - it is written that variance is time dependent.

Cut the crap

 
vladavd:

Why all this talk about SB if you're going to end up trading in the market anyway?

Because SB is a test case for any strategy.

If "patterns" and "waves" are identified on a number of SBs, then the price is 0

 
denis.eremin:

SB is a non-stationary process, just like the price series - it's written that the variance is time dependent.

Cut the crap.

You need to learn how to talk to people first, seizure.

 
vladavd:

Why all this talk about SB if you end up having to trade on the market anyway?

I have no idea. It's like everyone's gone crazy here...

 
Alexander_K2:

You need to learn how to talk to people first, you seizure.

Well, what can you do - I have to explain the basics of the theorist a hundred times, but I still don't get it.

 
Aleksey Nikolayev:


I've already written in this thread what the point of the comrade's actions is, the maths has nothing to do with it - it's just the end of the month... )

You may be right.

 
denis.eremin:

SB is a non-stationary process, just like the price series - it's written that the variance is time dependent.

Cut the crap.

If I understand correctly, Alexander is arguing that the behaviour of variance over time is in the nature of SB, and therefore the price series is more random than stationary SB. You argue that the variance is time dependent.

I prefer Alexander's position. What is your belief that the variance of a price series has any dependence on time based on, other than that there are time cycles of working periods.

 
Valeriy Yastremskiy:

If I understand correctly, Alexander argues that the behaviour of the variance over time is in the nature of SB, and therefore the price series is more random than stationary SB. You argue that the variance is time dependent.

I prefer Alexander's position. On what is your belief that the variance of a price series has any dependence on time, other than that there are time cycles of working periods.

1. What does "time cycles of working periods" have to do with it at all? What are these cycles - daily changes in volatility?

2. Break the price series into chunks, determine the variance of each chunk. Compare - if it is different - it depends on time.

 
Valeriy Yastremskiy:

the behaviour of the variance over time is in the nature of SB, and so the price series is more random than the stationary SB.

This is exactly the case.

If in SB the first differences are strictly stationary process, and the integrated series is stationary under identical samples or sets of realizations with finite sampling,

the price series does not have these properties.

Therefore, the price BP is much more complicated. There is nothing to argue about.

Reason: