04 August 2010

ECB, NFP and Conduits

I have mentioned what is called a long memory process in the market, but what is this. To me there are three great scientific works of recent time, special relativity, Godel's work on computability and Hurst's work. Of these Hurst is probably least known, but his work has most relevance to life.

Einstein's work is only relevant in extremes you simply will not experience in ordinary life (that is part of its brilliance, his deduction of the extraordinary from the ordinary). Godel's work is incredibly precise, but it is unclear what it actually means for computation as we experience it. But Hurst's work is relevant.

Hurst looked at data from the flow of the Nile and found correlations between recent data and past data. The structure of this as a process is a fading persistence (or long memory, but it is really a consequence of cyclical dynamic processes in nature). The question is, does this exist in the markets.

There is conflicting evidence for the stock market and stronger evidence it does not exist for forex. Hurst's work is tied up with chaos theory as the structure by which fading persistence is described is fractional in its dimensionality, it is like the structure in nature. But the strongest evidence for this kind of structure has been in commodities.

This is not entirely surprising. Commodities are highly valued by external factors, and subject to all sorts of inputs from nature (like seasonality). So this finding may not be saying anything at all. It is merely reflecting on inputs. What I am really saying here, is this kind of structure will not help you make any kind of predictive inference.

I have noted that the markets are to an extent not a product of nature, it is something new. Of these three markets, one can order them: commodities, stocks, forex in terms of the influence of nature in their structure. Why do I see it like this, normally people are seen as part of nature.

This is true in terms of our effect on nature and where we comes from, but it is the possibility we have a new kind of technology given by a process making natural structure, evolution (a fractal process itself). This is the work Penrose found intriguing.

This technology seems to give a capacity to lift ourselves away from natural computational processes, by virtue of the fact it is a different kind of computation, reflective of where Einstein saw beauty to the end of time rather than Godel or Hurst saw the fading beauty of the natural world or the limits of algorithms. It seems not unreasonable to me to assume that is in the markets.

What about Non Farm Payrolls ? Well, this one maybe is going to tell us whether the recovery is actually happening, strong enough to remove the possibility of a double dip recession. I believe the double dip is really only a figment of that conduit, it is the growing impossibility (the slide) to keep the illusion of asset deflation at bay.

But as I have said a recovery will clear all that. On the long term charts, USD/JPY, the bell-weather of a real recovery, has been testing that major support, but look at that movement up (on 15 min), that is a classic pattern of a computer driven surge trade.

Such a surge can have longer term significance as well, if the fundamentals support it or at least do not turn it off (it is saying well, let's go, USD is way too undervalued, but what fundamental support will it get on friday ?). A rise in USD has to work against the conduit, though, that is an issue until US interest rates rise and it disappears and that is what ECB is all about tomorrow (if $ interest rates are going to rise given a recovery, will ECB keep their rates above and maintain the conduit anyway ?).

From 1 min data behavior applied to long term charts USD/JPY is at crossroads, but the issue is has it just crossed it. Anyway that is one reason why forex is not like nature, because today the computer programs made a technical decision about USD. So does this mean stocks are more like the brain. My feeling is yes, but the possible advanced technology is more in forex.

Update: 8/5: Computer driven surges have a tendency to fizzle out, meaning popping to the top of RSI then drifting down. Presently USD/JPY has hit the bottom of RSI, it will be interesting to see what happens next.

© 2010 Guy Barry All Rights Reserved