Wednesday, November 2, 2011

2011.11.02 Interim DJIA Long $ Short Term
































You can see that there is a smaller ~ 3 to ~4 month cycle here superimposed on the larger set of cycles headed down. The large improvement in overall fit results because it appears to operate back to 1896 with some variations on the theme.

Visit econocasts.net for free sample trading model downloads. 

9 comments:

El Viejo said...

And more bad news from Europe and the US Q3 and still the market climbs higher. Amazing!

Paolo said...

Right, and as far as I can tell, functionally and structurally, the US is quickly approaching its own Minsky Moment

El Viejo said...

Interesting article on Risk in Science News Mag:

http://www.sciencenews.org/view/feature/id/335383/title/Beware_the_Long_Tail

Paolo said...

Thanks! Very cogent and a nice reminder of the pitfalls of modeling. It mentioned a few favorite scholars of mine. Mandelbrot being numero uno of course...meanwhile, talking about tails, I've got my DJIA cycle model taking a very long time to update itself with data from EOD 11/3. If it does not converge on a solution, it is toast. There is nothing subjective about it. There is some humor in this since the model may be chasing a tail of sorts.

Oliver said...

Paulo,

any update on how your model is going for the Dow Jones? Did it converge on a solution?

Regards

Oliver

Paolo said...

It converged on a solution after about six hours. Usually it takes a few minutes for EOD updates. I'm highly annoyed that I have not been keeping track of how many iterations it requires for convergence at EOD updates because I now suspect there may be a pattern to it; I can't go back to check.

El Viejo said...

If the implication in this video that the US govt is behind big US bank CDS on European banks is true then I would expect your model to diverge more as the inevitable is pushed out even further.
http://www.nakedcapitalism.com/2011/11/michael-hudson-on-the-showdown-in-gre

However if Reggie has it right US bank CDS may not be valid to a Greek 'default'.
http://www.zerohedge.com/contributed/entire-global-banking-industry-carrying

El Viejo said...

It's been my experience that the bigger the entity the longer it takes to react. There may be mini-cycles created in big events that cause reaction delays and stair steps up or down hill.

Paolo said...

El Viejo, you indirectly raise the question of why there are business cycles in the first place.

One factor suggesting human nature has not changed much is the extent of Fed/Gov meddling in the markets. Various schemes were proposed in the aftermath of 1929 which bear a resemblance to the current situation. I believe this is the definitive work on the subject, which was well worth the price of admission:

The Crash and Its Aftermath: A History of Securities Markets in the United States, 1929-1933 (Contributions in Economics and Economic History)