Likelihood ratio tests for model selection and non-nested hypotheses. (English) Zbl 0701.62106
The paper uses the Kullback-Leibler information criterion on measuring the acceptance of a model. A simple likelihood-ratio based on testing the null hypothesis that the competing models are equally close to the true data generating the process against the alternative hypothesis that one model is closer has been considered. The tests are classical in nature and the author proposes some directional symmetric tests for choosing between models. Some useful asymptotic properties are also derived.
Reviewer: P.W.A.Dayananda
MSC:
62P20 | Applications of statistics to economics |
62E20 | Asymptotic distribution theory in statistics |
62G10 | Nonparametric hypothesis testing |
62F03 | Parametric hypothesis testing |