Seminar in applied mathematics and statistics
SPEAKER: Mette Asmild (Department of Food and Resource Economics, University of Copenhagen)
TITLE: Statistical comparisons of production frontiers in nonparametric models
ABSTRACT: Non-parametric productivity analysis approaches, like Data Envelopment Analysis (DEA) are widely used in practice. However, their non-parametric nature, and the fact that the resulting efficiency scores are biased, makes statistical inference difficult. The few tests that are available in the literature are all based on asymptotic theory relying on large sample sizes, whereas situations with relatively small samples are often encountered in practical applications.
In this paper we propose three permutation tests: The first is a test for the hypothesis of constant returns to scale in DEA. The others are tests for general frontier differences and whether the production possibility sets are, in fact, nested. The advantages of permutation tests are that they are appropriate for small samples and furthermore have the correct size. Simulation studies show that the proposed tests do, indeed, have the correct size and furthermore higher power than the existing alternative tests based on asymptotic theory.
Based on joint work with Anders Rønn-Nielsen and Dorte Kronborg.
Thursday, May 23 at 14.15: Andrea Macrina
Friday, May 24 at 13.15: Amit Mitra
Friday, May 24 at 14.15: Sharmishtha Mitra
Wednesday, May 29 at 15.15: Frank van der Meulen