S&P 500 Poll

Thanks to everybody who took the time to vote in the most recent poll (located in the right hand column). The results (see below), as negative as they are, are about right in line with current market sentiment. So that leaves us with the million dollar question..."Where will the S&P 500 be on January 1, 2010?" There are a lot of estimates out there, most of which are higher than the current value, so how do we determine which estimate is correct? In my opinion, you'd have much better luck going to Vegas than trying to predict the value of the S&P 500 at years-end, however, there are a few ways that you can "nudge" the odds in your favor. An approach that I have been experimenting with uses a conditional probability model to forecast a "very rough" year-end estimate (link to Excel listed below). As of recently, there's been quite a bit of talk about a potential depression (assume dep'n = 10% unemployment, -10% GDP growth). Using a probability model, I can ask myself, "What is the probability of a depression?" and alternatively, "What is the probability of a Deep Recession?". By the way, I'm assuming we are already in the midst of a deep recession. As you can see from my model, P(depression) = .15 and P(remaining in a deep recession) = .85. For each scenario (recession and depression), I then predict an optimistic value and a pessimistic value given the prior outcome. For example, given a deep recession (85% chance), I came up with an optimistic value of 900 on the S&P by year-end and a pessimistic value of 750 (values are completely based on your outlook). Using probabilities, I can estimate the chance of the index reaching the optimistic value versus the index reaching the pessimistic value (by the way, we are assuming that these scenarios are mutually exclusive and exhaustive, so probabilities must sum to 1). Once you have your values and your probabilities, you can then just take the weighted average using your probabilities as your weights. The math will make more sense once you view the embedded equations. I encourage you to plug your own estimates into my model and see what you come up with. Just remember...this is a VERY rough estimate, but for a guy who enjoyed stats class, it serves as a basic tool for linking an economic forecast to a logical estimate for the S&P 500. As always, any comments or critiques are welcome.

Poll Results:

>1100 (33.3%)
900 (0%)
700 (66.6%)
<500 (0%)

To download the Excel file, please visit the link below and click on the Download tab located at the top, then choose Microsoft Excel.
http://www.scribd.com/doc/12966663/Conditional-Probability