UPDATE 9 Oct 2009: The method I'm using incorporates data from the market as it comes in. There might be a need to adjust the peak to the later half of Oct 2009. I'll keep you guys update if the calculation changes. I'll still call the peak before it happens.
Yes I know many consider it is an extremely futile activity trying to predict a top for the stock market. I've been wrong before - many times. But I keep having this delusion that I can "do it". Picking a top is the alchemy of finance - a preoccupation that financial pundits just get can't walk away from. I found something that would have worked for the past but don't know if it is completely useless for the future. But I did put in some amount of rigor and understand that the past may not be correlated to the future. If I'm wrong, don't look for me - nothing surprising since the stock market is considered by many to be a random walk. But if I manage to get it righ this time, give me one round of applause and remind to explain how it is done.
Here is the prediction:
The market will top out some time next week 12-15 Oct 2009 and go into correction that will last until early next year Feb-March 2010.