The effect of correlation coefficient among multiple input vectors on the efficiency mean in data envelopment analysis

In some of the papers on data envelopment analysis (DEA), there have been explained that if correlation coefficient between each pair of input (output) vectors is strong and positive, one of the input (output) vector could be omitted. The objective of this paper is to determine correlation coefficient threshold that beyond which omission of one or more input vectors have no statistically significant effect on the efficiency mean. The threshold identification in terms of some of the DEA models including CCR, CCRCSW, BCC and BCCCSW are performed. To analyze the data, analysis of variance (ANOVA) is used.