A Decision-Theoretic Formulation of Fisher’s Approach to Testing

摘要: In Fisher’s interpretation of statistical testing, a test is seen as ‘screening’ procedure; one either reports some scientific findings, or alternatively gives no firm conclusions. These choices differ fundamentally from hypothesis in the style Neyman and Pearson, which does not consider noncommittal response; tests are developed between two complementary hypotheses, typically labeled ‘null’ ‘alternative.’ The same presented typical Bayesian tests, where Bayes Factors used to judge relative support for null alternative model. this article, we use decision theory show that can also describe Fisher-style procedures, such approaches lead directly analogs Wald two-sided p-value, with frequentist properties be determined easily accurately. contrast these decisions do exhibit Lindley/Jeffreys paradox, tha...