Improving the efficiency measurement index using principal component analysis (PCA)

Abstract

This study aims at presenting an approach capable of improving the efficiency measurement index using the Principal Component Analysis. The main reason for the adoption of this approach in measurement process lies in the fact that during an analysis of data envelopment, there is often discrimination problems among measurement units, whether a measurement unit is efficient or not, especially there is a large number of variables (inputs and outputs). As to the size of a sample to be studied, and for addressing this problem, the Principal Component Analysis is adopted in this study, which is deemed one of the important statistical tools in decreasing original variable dimensions. For obtaining new variables representing principal components, factors for each variable have been identified, which works on determining its effect. Therefore, this study brings to light an integrated approach between two procedures: PCA-DEA, taking part in enhancing results of the efficiency measurement index. This approach has been applied to a problem related to the state financial management of the Iraqi Budget for the years (2005 – 2019). As per the decrease principal accomplished through the use of PCA, efficient years are identified.