No-decision classification: an alternative to testing

11 This paper proposes a new statistical technique for deciding which of two theories is better supported 12 by a given set of data while allowing for the possibility of drawing no conclusion at all. Procedurally 13 similar to the classical hypothesis test, the proposed technique features three, as opposed to two, 14 mutually exclusive data classifications: reject the null, reject the alternative, and no decision. Referred 15 to as No-decision classification (NDC), this technique requires users to supply a simple null and 16 a simple alternative hypothesis based on judgments concerning the smallest difference that can be 17 regarded as an economically substantive departure from the null. In contrast to the classical hypothesis 18 test, NDC allows users to control both Type I and Type II errors by specifying desired probabilities 19 for each. Thus, NDC integrates judgments about the economic significance of estimated magnitudes 20 and the shape of the loss function into a familiar procedural form. 21 © 2004 Published by Elsevier Inc. 22 JEL classification: C12; C44; B40; A14 23