This experimental model is an extension of the current model, with the addition of some routines that perform pattern matching at a higher resolution. Unfortunately, it is heavily CPU intensive, and I'm running a 48 CPU cluster, so I am not kidding when I say CPU intensive.
This takes the place of the usual Wednesday update since this week I cannot update Wednesday. So it is the first update of the Gregorian New Year.
It is always a bad sign when reality trends at 90 degrees to the model prediction. So - I think this week, either the DJIA breaks down, or the model breaks down. So far, the model for the DJIA converges quickly, so there is still hope. For the model that is.
The blog seems to get a lot of visitors from Germany, so I decided to see if I could work with the DAX and put up a DAX model chart. I am still processing it so it may be a few days and like the AMEX Oil Index, I may not get a chance to fully backtest it. It covers a short time period in relation to other models, but so far it looks reasonable.