Application of principal component analysis for parsimonious summarization of DEA inputs and/or outputs

Application of principal component analysis for parsimonious summarization of DEA inputs and/or outputs

0.00 Avg rating0 Votes
Article ID: iaor19982643
Country: Japan
Volume: 40
Issue: 4
Start Page Number: 466
End Page Number: 478
Publication Date: Dec 1997
Journal: Journal of the Operations Research Society of Japan
Authors: ,
Keywords: communications, programming: fractional, programming: linear
Abstract:

In Data Envelopment Analysis, when there are more inputs and outputs, there are more efficient Decision Making Units. Because usually inputs or outputs are correlated, they should be selected appropriately. We propose using principal component analysis as a means of weighting inputs and/or outputs and summarizing them parsimoniously rather than selecting them. In principal components (PCs), many weights for original variables may have negative values. A denominator of the objective function in fractional programming should be positive. In the basic model, a condition that the denominator must be positive is added. When the number of PCs is less than the number of original variables, a part of original information is neglected. In the modified model, a part of the neglected information is also used.

Reviews

Required fields are marked *. Your email address will not be published.