作者: Mitchell R. Watnik , Wesley O. Johnson
DOI:
关键词: Statistical hypothesis testing 、 Alternative hypothesis 、 Linear model 、 Algebraic number 、 Selection (genetic algorithm) 、 Mathematics 、 Statistics
摘要: SUMMARY. We consider the asymptotic behaviour of tests for non-nested linear model selection under alternative hypothesis. three testing procedures: J-test (cf. Davidson and MacKinnon, 1981); JA-test Fisher McAleer, modified Cox test Godfrey Pesaran, 1983). All are shown to have positive means true alternative. Thus, discriminating between models, they should be treated as one-sided tests. derive distributions slightly versions these statistics in order obtain relative eciencies. The selected since greater power than original statistics. Both algebraic theoretical relationships among indicated.