Pair trading and multicurrency arbitrage. The showdown. - page 110

 
Maxim Dmitrievsky #:
Co-integration in financial markets is another myth,
Is there no cointegration between the underlying asset and its futures?
 
Sergey Gridnev #:
And there is no cointegration between the underlying asset and its futures?
Well, give me an example.
There is no co-integration between near and underlying, it is one instrument in terms of prices, offset. There may be between the distant ones, but you will not be able to trade them properly due to lack of liquidity
 
Maxim Dmitrievsky #:
Well, give me an example.
Here on the forum stockbrokers have repeatedly given examples of trading.
 
Sergey Gridnev #:
Here on the forum, stockbrokers have repeatedly given examples of trading.
Is this a new-fashioned style of communication, writing for others? False very much
 
Maxim Dmitrievsky #:

Well forex is certainly a market, but not exactly a financial market, but more of an infobiz one

Stat arbitrage between indices, futures and other derivatives - we can conditionally call some of them cointegrated, but these instruments were specially created in advance, otherwise there would be no sense in them. And there the yields are small.

The link between the underlying and the secondary instrument should always be there, the returns there (between the underlying and the derivative) cannot be large, as well as losses, unless of course the underlying asset collapses/rises, it is not a calculation of a fair price relative to the real anything))). And for currencies, or the same indices, it's a no-go. Stocks, commodities locally, in general are also too voluminous and not unambiguous often. And derivatives from them of the first, or maximum of the second order, let's say))))) It is still possible to cope with it somehow)))))

 
Grigori.S.B #:

Rena's is the coolest. )))

If you've seen it, don't even remind me.

Let them look for it themselves.

 
Valeriy Yastremskiy #:

The link between the underlying and the secondary instrument should always be there, the returns there (between the underlying and the derivative) cannot be large, as well as losses, unless of course the underlying asset collapses/rises, it is not a calculation of fair price relative to the real anything)))) And for currencies, or the same indices, it's a no-go. Stocks, commodities locally, in general are also too voluminous and not unambiguous often. And derivatives from them of the first, or maximum of the second order, let's say))))) It is still possible to cope with it somehow)))))

Well, this connection is artificial, I would not consider these instruments as cointegrated, because they are essentially the same thing. And they can be traded through correlation as well :)

imho, of course

 
Maxim Dmitrievsky #:

Well this correlation is artificial, I would not consider these instruments as cointegrated, because they are essentially the same thing. And they can be traded through correlation as well :)

imho, of course

I don't understand a little bit about artificial correlation, a correlation (cointegration) is either there or it is not. Correlation is a confirmation of this connection, but not an absolute sign of connection. The holivar is not clear. It is clear that if there is a correlation, it is not bad to find why and the reasons for this correlation, but this is no longer techanalysis, this is the maze of real))))

maybe I don't know something))))

Of course, we can exaggerate that the constant property of linear relationship between series implies some causal relationship, maybe even not revealed, but the constancy hints that there is a relationship)))).

 
Valeriy Yastremskiy #:

I don't understand a bit about artificial connection, connection (cointegration) is either there or it is not. Correlation is a confirmation of this connection, but not an absolute sign of connection. The holivar is not clear. It is clear that if there is a correlation, it is not bad to find why and the reasons for this correlation, but this is no longer techanalysis, this is the maze of real))))

Maybe I don't know something)))

Of course, we can exaggerate that the constant property of linear relationship between series implies some causal relationship, maybe even not revealed, but the constancy hints that there is a relationship)))).

The significance of the notion of cointegration, when a derivative instrument repeats all the properties of the original one, but is slightly different, disappears. And so it is clear to everyone that the series are the same.

It is not necessary to build a z-score chart to trade stat arbitrage. You can do it in other ways.

The reverse is to take 2 stocks from the same sector and prove they are cointegrated. Or some macroeconomic indicators.
 

Valeriy Yastremskiy #:

One can of course exaggerate that the constant property of a linear relationship between series suggests some causal connection, maybe not even revealed, but the constancy hints that there is a connection))))

This is commonly referred to as multicollinearity. It is almost always present in the market and spoils life when making portfolios, that's why there are so many articles on this topic.

Cointegration is quite different, it allows you to get a portfolio with a stationary price.