What are the odds of 2006 being a positive year?
Here’s an interesting topic from Barry Ritholtz’s site, The Big Picture.
Link: What are the odds of 2006 being a positive year?.
I just had an interesting conversation with Mike Panzner, Director of Institutional Sales Trading at Rabo Securities. Mike is a quant who has an interesting take on the markets. Assuming the SP 500 index finishes about where the markets are at the moment …
Here is a great example of how even two good thinkers like Barry Ritholtz and his friend Mike can get confused by probability. I’m sure if you asked either of them the odds of a coin coming up heads for a fourth time AFTER you already had three consecutive "heads" they would both give the right answer.
Well this is virtually the same problem. If you think that the market is always pretty close to fairly valued, then the odds of an up year are 50%. Actually, because of the long-term upward bias the odds are better than that. The fact that this is the fourth consecutive year is utterly irrelevant unless you think that the last three years have sent stocks into overvalued territory.
So we are back to the question of fundamentals and underlying value, not any sophisticated "quant" look at history or probability. Interestingly, none of the commenters on Barry’s site point this out.
The market implication? To the extent that current prices reflect the thinking of people like Barry and his readers, current prices are a bargain.
The key difference is that coin tosses are a 50/50 affair — regardless of the streak, they remain a random event.
Markets, are the other hand, are not “True.” They exhibit specific tendencies that reflect non random characterstics, such as persistence, sensitivity to initial conditions, and especially, cylicality.
In particular, the 4 year cycle that markets exhibit is fairly established, with highs in the 4th year of a President’s term, and lows in a the 2nd year. (See this for more details:
http://bigpicture.typepad.com/comments/2005/09/presidential_cy.html )
The lack of 4 year streaks, excepting two atypical eras, is consistent with this phenomena — and its why random chance (i.e., flipping a coin) yields a very different behavior than do markets . . .
Thanks for your comment, Barry.
Your original post, after listing the streaks, said: “Under the circumstances, that implies pretty good odds that 2006 is less than likely to be an up year, just based uon historical averages.”
We all understand that years may not be completely independent, although the direction of the dependence (trend persistence) seems counter to your point.
Try this. Tell us what you think the odds were of a four-year streak BEFORE IT STARTED, in Dec 2003. What do you think the odds are now that three years have been booked?
Whether you are right on the market forecast remains to be seen, but I don’t think it has anything to do with streak analysis.