Further Evidence on Seasonal Adjustment of Time Series Data

Abstract

The purpose of this paper is to provide evidence that the Bureau of the Census' X–ll program for seasonal adjustment [3] overstates the incidence of seasonality in some forms of times series data. This problem arises in a recent study by Bonin and Moses [1] (hereafter B-M) indicating that 7 of the 30 Dow Jones Industrial stocks exhibited persistent seasonal patterns during the period July 1962 through June 1971.