Model Recognition by Using Principal Component Analysis (PCA) Approach

In this paper, an alternative model recognition method is proposed by using Principal Component Analysis (PCA). This alternative approach is used to choose the optimum model for fitting the index of real compensation per hour (Y) and labor productivity per hour (X) in the business sector of the U. S. economy for the period 1960-1991. Comparison is then made with the existing methods such as ranks of the R-2, Adjusted R-2(R-2), Akaike Information Criterion (AIC) and Schwarz's Information Criterion (SIC) values. The empirical evidence shows that the proposed method has the same ability to choose the best fitted models. The main attraction of this method is that it can be applied to all types of data scale; however, the existing methods not work for all types of data scale. Additionally, the proposed method has a clear edge over its rival because the PCA uses actual observations. Hence, we suggest to use the proposed method instead of the existing methods in determining the best fitted model.