High volatility

The extreme trend conditions of the last few days were eased substantially last night.  The Beta chart shows how much air was let out of the balloon - especially in the AUD crosses.  This in turn resulted in the Detma model taking a very large loss on paper on the day.

Systematic models like Detma, by their very nature, utilise a consistent methodology over time.  This results in their replicating the function of a filter as it were.  That is, on top you get a whole bunch of price activity which is filtered, or acted upon, in a consistent manner by the model.  Out the other end comes the performance results.  An analysis of these performance results will yield certain risk reward characteristics which can tell us a lot about how the market is behaving at any point in time.

I mentioned yesterday that Detma was up 10.2% month to date.  However last night Detma gave back an astonishing 5.6%.  I was so surprised by the extent of this retracement that I did a bit of analysis on the results for the currency portfolio that Detma trades.  The table at left summarises the results. 

Essentially, since I started trading the model on 1 September 1995 to yesterday, 27 May 2010, there have been a total of 3,845 daily performance observations.  Of these only 5 have been losses of -5.6% or more.  So yesterday's result had a probability of occurrence of only 0.13% - a black swan event by any name.

In fact if you look at the table closely you will see that the entire history of the model had 18 black swan event losses (I'm defining a black swan as an event outside of the 99% level of confidence i.e. 3 stdev) and 29 black swan event gains.  Of the 18 loss events, 12 (or 66%) have occurred in the last 18 months since October 2008.  now if that doesn't tell you something about the market nothing will.  Bear in mind as I mentioned above, Detma has traded these currencies in the same way since inception.  So it's not the model that has changed - it's the market. 

The first thing that would need to happen if the Detma model were live and running money at the moment would be that the gearing would be maintained much lower than the typical 5 times.  Secondly I would be reviewing the entire research that supports the philosophy of Detma.  Yes markets definitely do trend but why the extremes in price action.  Should the MTI measure be allowed to take more than 40% off the table during trend extremes?  Many questions need to be addressed by systematic models at the present.  More on this later.

Is the AUDUSD completing a 4th wave?  Instead of a triangle its has done an ABC flat correction.  What are the stock markets going to do from here?  They seem to be leading the way. 

I'm short AUDUSD at .8500.  I won't be stubborn but if it does hold around here it could be the start of the 5th wave down to 0.7500.

Frank