Open Conference Systems, 50th Scientific meeting of the Italian Statistical Society

Font Size: 
An evaluation of KL-optimum designs to discriminate between rival copula models
Silvia Angela Osmetti

Last modified: 2018-05-17

Abstract


The problem of model discrimination has prompted a great amount of research over last years. According to the specific characteristics of the rival models (nested, non-nested, linear or not) different optimum criteria have been proposed to select design points with the aim to discriminate between rival models. Ds-, Tand KL-criteria are the most known. Up to our knowledge, in the literature there is not any study to evaluate the performance of these discrimination criteria. In this work, via a simulation study and focusing on rival copula models, we analyze the performance of the KL-optimum design applying the likelihood ratio test for nonnested models.

References


1. Atkinson, A. C., Cox, D.R. : Planning experiments for discriminating between models. J.R. Statist. Soc. B, 36, 321–348 (1974)

2. Atkinson A. C. and Fedorov V. V.: The Design of Experiments for Discriminating Between two Rival Models. Biometrika, 62, 57–70 (1975)

3. Cox, D.R.: Tests of separate families of hypotheses, In: Proceedings of the Fourth Berkeley Symposium on Mathematical Statistic and Probability, University of California Press: Berkeley, 105–123 (1961)

4. Cox, D.R.: Further results on tests of separate families of hypotheses.Journal of the Royal Statistical Society B, 24, 406–424 (1962)

5. Deldossi, L. and Osmetti, S. A. and Tommasi, C.: PKL-Optimality Criterion in Copula Models for efficacy-toxicity response, In: mODa 11 - Advances in Model-Oriented Design and Analysis, Kunert, J., Muller, C.H., Atkinson, A.C. (eds), Springer International Publishing: Heidelberg, 79–86 (2016)

6. Lopez-Fidalgo, L.J., Tommasi, C., Trandafir, P.C.: An optimal experimental design criterion ´ for discriminating between non-Normal models. Journal of the Royal Statistical Society B 69(2), 231–242 (2007)

7. Nelsen, R.B., An Introduction to Copulas. Springer, New York (2006)

8. Pesaran, H. and Weeks, M.: Non-nested Hypothesis Testing: An Overview. In: A Companion to Theoretical Econometrics,BH Baltagi (eds), 279–309 (2001).

9. Perrone, E. and Muller, W.G.: Optimal designs for copula models. Statistics, ¨ 50, 917–929, (2016)


Full Text: PDF