Prediction of A CRS Frontier Function and A Transformation Function for A CCR DEA Using EMBEDED PCA

Data Envelopment Analysis is a nonparametric tool for measuring the performance of a number of homogenous Decision Making Units. In this paper, Principal Component Analysis is used as an alternative tool to estimate the frontier in a Data Envelopment Analysis under the assumption of Constant Return to Scale. Apart from this, in the context of a multiple inputs and single output, a transformation function, is developed here using the Most Productive Scale Size condition stated by Starrett. This function complies with all postulates of a frontier function and is very similar to the formula given by Aigner and Chu. Moreover, it is capable of defining the threshold value for any resource.