IMPACTS OF ADVANCES IN INFORMATION SCIENCE ON BUSINESS FORECASTING
Technology management is assuming greater importance in contemporary business organizations (Cyert & Kumar, 1994, p. 333). Among other organizational benefits of the application of technological innovations in information science (IS) is an enhanced capability to perform business forecasts. This research examines the impacts of advances in IS on business forecasting.
A forecast is an estimate of the future value, or future values, of a variable (Webster, 1994, p. 6). To be meaningful, forecasts must project values of a variable in measurable units of the variable. A number of procedures exist by which forecasts may be developed. The best forecasts are those developed through the application of statistical procedures. Among the major statistical forecasting procedures are the following:
2. Cyclic demand projections, which consider random variations, and which require the use of multiple regression equations.
3. Curvilinear regression projections, which consider multiple relationships among many variables.
The simplest of the quantitative techniques which may be used in forecasting is the fitting of a trend line. The major problem with this methodology is that it relies almost entirely on past performance. The fitting of a curve through regression analysis is methodologically similar to trend analysis. The curve fitting regression equation, however, permits the introduction of new data into the analysis, through the refinement of projections for the independent variables involved.
Curve fitting may be accomplished through the application of either bivariate analysis, or by extending bivariate analysis to multiple regression analysis. Multiple regression analysis should be employed, if forecasting accuracy and validity are improved through the inclusion of additional independent variables.
Fitting a simple equation to a time series is the ...