Estimating semi-parametric output distance functions with neural-based reduced form equations using LIML
Efficiency analysis is an important tool for evaluating firms' performance. This paper introduces a novel approach for measuring technical efficiency (TE) in the case of technologies with multiple outputs which deals with the endogeneity of outputs issue. The proposed approach uses Artificial Neural Networks (ANNs) and the method of Limited Information Maximum Likelihood (LIML). The validity of the proposed approach is illustrated by fitting it to a large US data set for all commercial banks in the 1989-2000 time span. Meanwhile, we compare the proposed approach to the single-equation Translog output distance function and the proposed approach was found to yield very satisfactory results, while dealing with the issue of the endogeneity of outputs.
Year of publication: |
2010
|
---|---|
Authors: | Vouldis, Angelos T. ; Michaelides, Panayotis G. ; Tsionas, Efthymios G. |
Published in: |
Economic Modelling. - Elsevier, ISSN 0264-9993. - Vol. 27.2010, 3, p. 697-704
|
Publisher: |
Elsevier |
Keywords: | Output distance function Translog Technical efficiency ANN LIML |
Saved in:
Online Resource
Saved in favorites
Similar items by person
-
Vouldis, Angelos T., (2010)
-
Globally flexible functional forms : the neural distance function
Michaelides, Panayotis G., (2010)
-
Michaelides, Panayotis G., (2011)
- More ...