You are not currently logged in.
Access JSTOR through your library or other institution:
Random Effects Logistic Regression Model for Data Envelopment Analysis with Correlated Decision Making Units
S. Y. Sohn and H. Choi
The Journal of the Operational Research Society
Vol. 57, No. 5 (May, 2006), pp. 552-560
Stable URL: http://www.jstor.org/stable/4102452
Page Count: 9
Preview not available
This study attempts to exploit information from environmental variables together with data envelopment analysis (DEA) efficiency scores for efficiency predictions of groups with more limited information. Based on DEA efficiency scores, decision-making units (DMUs) are sorted into two sets containing efficient and inefficient units, respectively. Then they are reshuffled into homogeneous groups with respect to environmental factors. We assume that the efficiency of DMUs in such a homogeneous group would be correlated. However, efficiency of different groups would vary. A beta binomial logistic model is proposed to fit such phenomena and is applied to predict the performance of a new group of commercialization projects for given environmental characteristics.
The Journal of the Operational Research Society © 2006 Operational Research Society