This research provides a critique of trend analysis. Trend analysis is one approach to time series projection.1 Time series projection may be as uncomplicated as the projec tion of an unadjusted trend, or it may be based upon more sophisticated adjustments for seasonal and cyclical patterns.2 In either case, or in some middle ground case, such as the application of least squares curve fitting (regression) to time series data, to project future values of a variable, a future trend is projected.3 The primary concern of this research, however, is basic trend analysis, which includes (1) arithmetic trend, (2) moving average trend, (3) exponential smoothing, and (4) exponential trend.4 One so called middle ground approach to trend analysis the application of least squares curve fit ting (regression) to time series data is also considered, because of its wide use in conjunction with each of the four approaches to trend projection indicated above.5 Regression
1J. L. Pappas, and E. F. Brigham, Fundamentals of Manager ial Economics, 3rd ed. (Chicago: The Dryden Press, 1989), 138.
2J. J. McAuley, Economic Forecasting for Business (Engle wood Cliffs, New Jersey: Prentice Hall, Inc., 1986), 10 31.
3R. J. Tersine, Production/Operations Management, 3rd ed. (New York: North Holland Publishing, 1990), 431 433.
4G. W. Summers, W. S. Peters, and C. P. Armstrong, Basic Statistics in Business and Economics, 5th ed. (Belmont, Cali fornia: Wadsworth Publishing Company, 1989), 557 565; Tersine, 432 444.
5R. C. Pfaffenberger, and J. H. Patterson, Statistical Methods for Business and Economics, 3rd ed. (Homewood, Illinois: Richard D. Irwin, Inc., 1987), 859 865.
analysis may be either simple (one explanatory variable), or multiple (more than one explanatory variable).
TREND ANALYSIS: APPLICATION, PROCEDURES, AND CRITIQUE
Managers and administrators of organizations use statis tics in three general ways. These three ways are as...