Custom Search

Quite frankly, it is difficult to establish a neat and clean argument for the eight-year pattern based only on actual political or economic factors. Using Figure 3's presentation, the pattern must be explained as six average years and two strong years. One answer might involve market valuation; after six average years of price appreciation, the natural growth of earnings, dividends and book values could cause the market to reach undervalued levels.

This sets the scene for the two strong years, which in turn leads to an overvalued market. Overvaluation then sets up on the six average years of price appreciation, and the pattern then repeats. The reader is left to judge the merits of this or similar arguments.

Many sophisticated ways exist to probe data relationships. Here, however, are two simple and intuitive approaches that utilize standard deviation. In an ideal world, a historical pattern would be perfectly strong and uniform, repeating itself over and over. In such a case, the several annual returns that make up each year of the pattern would be identical. The standard deviation of each column of annual returns would therefore be zero.

Conversely, when a pattern is weak and the returns of the years that make up each year of the pattern vary over a wide range, the standard deviation would be high. Thus, we can use the standard deviation of the years that make up each year of the pattern as a measure for the historical consistency of the pattern. The higher the standard deviation, the less consistent that pattern has been, and the lower the standard deviation, the more consistent.

Going a step further, how might the standard deviations of the actual pattern compare with a sampling of the same data divided into four types of years at random? If the Presidential election effect actually has an impact on the data, its standard deviation should be less than the standard deviation of the same data ordered at random.