12 month % change in the S&P500 by percentile, 1945- May, 2007, excluding dividends (all rolling 12 month periods), n = 700 :

The S&P500 Itself: Descriptive Statistics

In order to determine whether a given independent variable (such as a change in earnings or interest rates) identifies above or below average subsequent periods of performance in the S&P500, we must first determine what the normal range of values in the S&P500 have been historically. It is useful to break down all rolling (overlapping) 12 month S&P500 percentage changes (excluding dividends) to get some idea of what "normal" is.

12 month % change in the S&P500 by percentile, 1945- May, 2007, excluding dividends (all rolling 12 month periods), n = 700 :

	Percentile:
Max:	53.4%
90%	28.6%
75%	19.6%
50%	9.5%
25%	-1.9%
10%	-12.6%
Min:	-41.4%
Mean:	8.4%
Standard Deviation:	15.7%

During the 700 months of this observation period, the S&P 500 gained an average 8.4 percent per year. Note that the market gained 53.4% maximum and lost 41.4% during its worst 12 month period. 50% of the time the market rose 9.5% or more. Only 25% of the time did it rise greater than 19.6%, and only 10% of the time did it rise more than 28.6%. It is reassuring to note that 75% of the time the market's performance was better than -1.9%.

Note that the standard deviation of market returns was 15.7%, that this statistic should be used with great caution since it implies a normal distribution. If the market returns were normally distributed, then the 41.4% maximum 12-month decline should occur only once every 110.8 years, but since it occurred during our 60-year sample, we get some idea of how fat the tails (or extreme ends) of our distribution are.