So, your criticism is the day to day forecast? I don't do much of that, typically (when I do, it mostly in response to thread's participants queries as to my opinion), I venture to try and identify short term moves within the broad, often, yearly, model. Sure, sometimes, things don't work (two very bad periods were the 2000 Nov/Dec, and mid Nov/Dec. of 2001, both of them , IMTO, due to unforeseen events, the former, the undecided election, the latter, the much more rapid than expected collapse of the Taliban), but they do more than 60% of the time, and that is quite enough (with an appropriate SL policy) to do quite nicely. As for the Q and few others, they are traded both in up and down channels, mostly, because I do not short. Larry's model (the old one) was whipsawed about three times in the May/Sep time frame of last year as well. With the new model, going long will occur only when the fed model shows stock to be cheap (I presume in conjunction with a trend change). It is a very good model, yet, it did not warn of last year's February massacre, the April call missed the first 30% (I don't remember if he had an early January call but it would have to be after January 4th, and thus be whipsawed as well by the February decline), it works best in long trending markets, not in the volatile market we are having and probably going to have. My bet is that if we get the late June decline that my current "yearly" forecast has, Larry's model will flash a buy signal around the fourth of July. As far as I am concerned, I'll be happy if the turnips will be as bad as they were in the last two years in the next two years (g). Saying that no one gets it right, is not accurate, unless your "right" is 100% right.
Zeev |
