From theory to practice. Part 2 - page 108

 

Alexander_K2:

Look further for differences from SB.

Aleksey Nikolayev:

Yes, I will do so.

Well, by and large it's almost the only way to earn on a single linear instrument (I leave in brackets all sorts of exotics: HFT, etc.).

 
Доктор:

Well, by and large this is almost the only way to make money on a single linear instrument (I leave in brackets all the exotics: HFT, etc.).

On a portfolio, too, if you think of multi-dimensional SB (multi-dimensional Wiener process, for example). A typical example is trying to make the portfolio price stationary, which is obviously impossible if prices are multivariate SB.

With HFT there is a significant complication in the sense that some process control appears. This is already a much more complex matstat if one tries to do it scientifically.

 
Доктор:

Well, by and large this is pretty much the only way to make money on a single line instrument (I leave in brackets all the exotics: HFT, etc.).

Not the only one.

When developing my own TS, I did not and do not make any comparison with SB at all. I'm totally indifferent to it. There was my thesis and books/monographs by Shelepin L.A. Everything. Then there was communication with people, who are just as absolutely far from mathematical exercises, but close to physics and biology.

My results may not be great, but I don't make making money in the market an end in itself. It is just a leisure time activity, a hobby. An interesting hobby. Fascinating.

 
secret:

So fluctuations with a non-permanent period are cyclical. Ideologically, they are not that difficult to "periodise". Technically it is more difficult.)

It is possible to try, of course, but for some reason the joke about streaks in life comes to mind - "it turns out it was a white streak")

 
Aleksey Nikolayev:

On a portfolio, too, if you think of multivariate SB (a multivariate Wiener process, for example). A typical example is trying to make the price of the portfolio stationary, which is obviously impossible if prices are multivariate SB.

With HFT there is a significant complication in the sense that some process control appears. This is already a much more complex matstat if one tries to do it scientifically.

As for HFT, especially exchange trading, the mathematics there is primitive, and the edge there is achieved at the expense of infrastructure. But the formulated rule (about earning on differences from SB) forHFT is also fulfilled: on sub-second intervals Hurst is noticeably less than 0.5.

 
А! I forgot to mention Gunn... How come... Unforgivable. The Supreme Master! The fundamentals of market cyclicality are, of course, to his credit.
 
Alexander_K2:

Not the only one.

In developing my own TS, I did not and do not make any comparison with SB at all.

You obviously do. The large values of your "magic indicator" are obvious points of rejection of the SB hypothesis. That you do not understand this is your problem, failure and shame.

 
Alexander_K2:

Not the only one.

In developing my own TS, I did not and do not make any comparison with SB at all. I am totally indifferent to it. There was my diploma thesis and Shelepin's books/monographs. Then there was communication with people, who are just as absolutely far from mathematical exercises, but close to physics and biology.

My results may not be great, but I don't make making money in the market an end in itself. It is just a leisure activity, a hobby. An interesting hobby. A fascinating one.

Almost the only one. And as for your path in trading, not everyone gets into deep theory or uses it. But that doesn't negate the fact that the theory is valid.

 
Alexander_K2:

Mind your own business. I've been haunted by this idiot for four years. Some idiot who should be put in an asylum.

Someone should warn people who are suffering that they are being sold nonsense)
 
Доктор:

As for HFT, especially exchange-traded HFT, the maths is primitive and the edge is achieved by infrastructure. But the formulated rule (about earning on differences from SB) forHFT is also fulfilled: on subsecond intervals Hurst is noticeably less than 0.5.

As I see it, HFT is, first of all, frontrunning. I don't particularly distinguish scalping (from a theoretical point of view) from regular speculation. I agree that on small scales prices are rather persistent and even advised the topicstarter to use this to diversify and reduce the drawdown of his return system.

Reason: