|  SEP 1986

Runs: A method for detecting 'rhythms' in prices by Clifford J. Sherry, Ph.D.

Runs: A method for detecting 'rhythms' in prices by Clifford J. Sherry, Ph.D. Do runs or clusterings of like observations, such as 'rhythms' of price increases or decreases, occur by chance alone or is there some deterministic process at work? Statisticians have provided us with several simple methods for determining if the number of runs (clusterings) in a set of data (like the prices of stocks or commodities) occurs by chance or not. The data used to illustrate these methods are shown in Figure 1. These data were abstracted from a longer series (Closing prices, soybean meal, 4/1/80—4/30/81) that has been used to illustrate other statistical methods in technical analysis. In the first method (Stocks & Commodities, October, 1985), we examined the sequential pairs of prices and determined if the first price pair was larger or smaller than the second and recorded a – or a + respectively. Then we compared the second and third price, the third and fourth, and so on. The results of these comparisons for the months of April, May, and June, 1980 are shown in the table under Method 1.

by Clifford J. Sherry, Ph.D.