Title :
Estimation in DEA by possibility regression with the upper and lower approximations
Author :
Entani, Tomoe ; Ichihashi, Hidetomo ; Tanaka, Hideo
Author_Institution :
Graduate Sch. of Eng., Osaka Prefectural Univ., Sakai, Japan
Abstract :
The exponential possibility regression is applied to estimate interval efficiency values. The interval efficiency value is obtained by interval DEA. It shows the possible efficiency and its upper and lower limits are obtained from the optimistic and pessimistic viewpoints for an object to be analyzed. The interval efficiency value for the object is calculated relatively by using all given inputs and outputs without assuming an identical single model for all DMUs. We analyze the interval efficiency values by regression analysis so that the identical model can be obtained. The obtained linear model in regression analysis can represent the relation between the given inputs and outputs and the obtained interval efficiency values in interval DEA. When new objects are added, we can estimate their interval efficiency values using the models obtained by regression analysis
Keywords :
data envelopment analysis; decision tables; possibility theory; statistical analysis; DEA; data envelopment analysis; decision making units; exponential possibility regression; interval efficiency values; linear model; nonparametric technique; regression analysis; Data envelopment analysis; Decision making; Educational institutions; Engineering management; Gaussian distribution; Information management; Input variables; Linear systems; Regression analysis; Upper bound;
Conference_Titel :
IFSA World Congress and 20th NAFIPS International Conference, 2001. Joint 9th
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-7078-3
DOI :
10.1109/NAFIPS.2001.943654
