goodness-of-fit test

todo

Philosophically, goodness-of-fit tests are interesting, because one usually does not have a specific alternative in mind. They are thus in violation of the law of likelihood and the likelihood principle. In general they are contrary to claims that evidence must be measured against a particular alternative (see also evidence against the null). They are thus somewhat anathema in the Neyman-Pearson paradigm (though Neyman and Pearson didn’t talk in terms of measures of evidence, they talked about decision-making).