Dear Thread,
I'd like to talk about basket design for a moment. If you look at any particular chart of the cluster indicator, you will initially detect what looks like randomness: a mass of undulating lines. These are, however, as we know, the ebb and flow, the amplitudes and distortions, of the forex world. There is nothing---no line here---that is not related to every other line through a multitude of factors.
A plethora of factors, frankly, in their range and complexity. A veritable riot list of factors truly beyond our ken. And so, as far as we traders are concerned, what we're facing is a type of "controlled" randomness. I say, "control" because there does exist a fundamental relationship between these lines. And that "control" and that randomness can, I believe, be captured profitably through basket trading in the sense that basket trading allows you to transcend the noise of the market with its scattershot approach. Jokingly, you could even call basket trading "group therapy" for currency pairs since they are all present in a portfolio to help each other out. But to capture the randomness without the control is dangerous. It is that "control" which puts a floor on the basket's risk. To say it in simpler words, what you want to do is to be able to buy a basket of currency pairs in this form of stat arb which will allow you to hang in there, time-wise, long enough for the basket to swing in your favor before you get stopped out. Which in turn means that you have to be able to capture the general "rules" of that randomness. How can that be done? In the spirit of less is more, what you see on the chart are three basic currency movements:convergence of the lines from extremities, divergence of the lines from the 0 line (or thereabouts), and cross-over/unders of the lines wherever they occur, i.e. relative value plays. Hence, it should most always be possible to construct unique, "designer" baskets based on these three observations that will mitigate risk and enhance return. What would then add to the RR of any particular basket would be to consider swap rates as part of the design as well as considering a "hedged" function based on correlation and vols---but that's enough for now. Appreciate all thoughts.
Thanks very much for your attention.
Lenoxer
I'd like to talk about basket design for a moment. If you look at any particular chart of the cluster indicator, you will initially detect what looks like randomness: a mass of undulating lines. These are, however, as we know, the ebb and flow, the amplitudes and distortions, of the forex world. There is nothing---no line here---that is not related to every other line through a multitude of factors.
A plethora of factors, frankly, in their range and complexity. A veritable riot list of factors truly beyond our ken. And so, as far as we traders are concerned, what we're facing is a type of "controlled" randomness. I say, "control" because there does exist a fundamental relationship between these lines. And that "control" and that randomness can, I believe, be captured profitably through basket trading in the sense that basket trading allows you to transcend the noise of the market with its scattershot approach. Jokingly, you could even call basket trading "group therapy" for currency pairs since they are all present in a portfolio to help each other out. But to capture the randomness without the control is dangerous. It is that "control" which puts a floor on the basket's risk. To say it in simpler words, what you want to do is to be able to buy a basket of currency pairs in this form of stat arb which will allow you to hang in there, time-wise, long enough for the basket to swing in your favor before you get stopped out. Which in turn means that you have to be able to capture the general "rules" of that randomness. How can that be done? In the spirit of less is more, what you see on the chart are three basic currency movements:convergence of the lines from extremities, divergence of the lines from the 0 line (or thereabouts), and cross-over/unders of the lines wherever they occur, i.e. relative value plays. Hence, it should most always be possible to construct unique, "designer" baskets based on these three observations that will mitigate risk and enhance return. What would then add to the RR of any particular basket would be to consider swap rates as part of the design as well as considering a "hedged" function based on correlation and vols---but that's enough for now. Appreciate all thoughts.
Thanks very much for your attention.
Lenoxer