Hypothesis Testing

Criminal cases are continually hinging upon that one point. A man is suspected of a crime months perhaps after it has been committed. His linen or clothes are examined, and brownish stains discovered upon them. Are they blood stains, or mud stains, or rust stains, or fruit stains, or what are they? That is a question which has puzzled many an expert, and why? Because there was no reliable test. Now we have the Sherlock Holmes’ test, and there will no longer be any difficulty. Sherlock Holmes in “Study in Scarlet” A first step in data modeling and understanding is the estimation of parameters in a supposed model. The second step—and very important statistical work—is the inferential conclusion on a hypothesis of the involved parameters. The construction of tests for different kinds of hypotheses is at the heart of this chapter.