TY - JOUR
T1 - The effect of correlation coefficient among multiple input vectors on the efficiency mean in data envelopment analysis
AU - Farzipoor Sean, R.
AU - Memariani, A.
AU - Lotfi, F. Hosseinzadeh
PY - 2005/3/15
Y1 - 2005/3/15
N2 - 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.
AB - 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.
KW - Analysis of variance
KW - Correlation coefficient
KW - Data envelopment analysis
UR - http://www.scopus.com/inward/record.url?scp=10444235093&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=10444235093&partnerID=8YFLogxK
U2 - 10.1016/j.amc.2003.12.117
DO - 10.1016/j.amc.2003.12.117
M3 - Article
AN - SCOPUS:10444235093
SN - 0096-3003
VL - 162
SP - 503
EP - 521
JO - Applied Mathematics and Computation
JF - Applied Mathematics and Computation
IS - 2
ER -