A part of statistical inference; an algorithm for addressing question of interest, usually put forward by a theory. A typical procedure consists of (1) setting the null hypothesis and the alternative hypothesis (i.e. formulating the question as a simple claim and the refutation of this claim), as well as the level of significance (the probability of type I error); (2) choosing an appropriate test statistic (a quantity whose distribution depends on whether the claim is true); (3) establishing the distribution of the test statistic under the assumption that the null hypothesis is true; (4) computing the value of the statistic from the data and making the decision to reject or not to reject the null hypothesis, usually by comparing the value of the test statistic with the ‘critical values of the distribution under the null hypothesis. Alternatively, the p-value of the test statistic can be used to determine the probability of making a mistake by rejecting the null hypothesis.
|Reference: Oxford Press Dictonary of Economics, 5th edt.|