Rob's comments made me think some here who aren't on the pk list might be interested in this. ------------------------------------------------------------------------- From: Trond Andresen [mailto:[EMAIL PROTECTED]] The dramatic fall of the NASDAQ to a two year low the last days may be a forewarning of more than a recession. In September 98 we had a seminar around a stock market model of mine, see my opening message for the pkt seminar, http://csf.colorado.edu/cgi-bin/mfs/55/csf/web/mail/pkt/sep98/0277.html See also the paper on http://csf.Colorado.EDU/authors/Andresen.Trond/stock-market.pdf It seems to me that the behavior of the market since 1998 is a confirmation of the dynamics predicted in my model. The main behavior demonstrated is a decade-long cycle. The length of one period is basically decided by the time for moods (optimist or pessimist) to propagate among agents. Moods are self-reinforcing both on the upswing and downswing, since mood-driven buying or selling amplifies the current mood via price change. At the same time there is a ceiling and a floor for index levels. The ceiling (which obviously was reached last year) is there because P/E-ratios become very small so that an even stronger belief in even stronger price appreciation becomes neccessary for an agent to continue buying. But if they do, P/E-ratios become even smaller and still stronger price appreciation must be demanded to compensate for this. Some agents see that this is an unsolvable scenario and decides that the good times have culminated. They shift to a bearish trading mood. A larger occurence of selling, even if indices are still on the rise because the main mood is still optimism, leads to the appreciation rate starting to weaken. Since this happens in a situation where agents demand a stronger appreciation rate than ever because their earnings are negligible or zero, this will lead to more agents understanding the impossibility of this, and shifting to a selling mood. The self-reinforcing process from index rise, flattening out, and later starting to slide, is mercilessly in motion. Note that we do not need any panic or crash for this to happen. It is a smooth limit cycle, to use dynamical systems terminology. It occurs beacuse the model essentially consists of unstable up- and downswings bounded by a ceiling and a floor. Any panics come on top of these dynamics. This sluggish, very-large-inertia process is not something stopped by small interest rate cuts. I have yet to find related models of long-term stock market dynamics in the literature. What i have found concerns itself solely with modeling and explanation of short-term excursions, small bursts and temporary instabilities around some "efficient" price. I think this is strange. But if there are models I am not aware of, please inform me. Trond Andresen
