Abstract
The results of correlation analysis are often intended for the eyes of the non-statistician as well as for those of the statistician. The problem, then, is to furnish a clear general picture and no more than that. Unnecessary detail tends to provoke understandable impatience. The conventional solution' is a tabulation of estimated values calculated from the regression equation, showing various combinations of the independent variables. But even the best of tables is uninviting reading matter and, for most readers, is not as easy to interpret as a graphic portrayal of the same data. A simple method of graphic presentation is shown here. It is an application of the familiar band or strata chart. The estimate is shown as the sum of the weighted values of the components from which it is made. By this method the reader intuitively grasps the very natural idea of several forces combining to determine the estimate. It is also easy to observe the closeness, or lack of closeness, of the estimated figures to the actual. The extent to which the actual and estimated series correspond graphically, clearly indicates the degree of correlation. Arrangement of the data as a time series is advantageous not only for the possible light that may be thrown on year to year changes in the data, but also because it automatically answers the typical question: What happened in that year? The method of preparation is simple. An equation showing the relationship between the dependent series and those with which it is correlated is a by-product of the ordinary correlation analysis. In the illustration, the equation was of the following form: Shipments of lumber = A constant + (a weight X value of new private construction) + (a weight X value of new public construction) + (a weight X personal consumption expenditure for furniture). The value series used have been deflated by building materials and house-furnishings price indexes. The constant, which is a balancing factor in the equation, gives the height of the lowest band. The height of the band for private construction is equal to that series multiplied by the appropriate weight from the equation. The two remaining bands are plotted in the same way. The result is that the total height of the four bands adds to the estimate of lumber shipments. By the use of this chart, the influences of the changes in private and public construction and furniture consumption on lumber shipments may be readily seen.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.