ABSTRACT
In the general one-sample problem, the available data consist of a single set of observations, usually a random sample, from a cdf FX on which inferences can be based regarding some aspect of FX . The tests for randomness in Chapter 3 relate to inferences about a property of the joint probability distribution of a set of sample observations which are identically distributed but possibly dependent, i.e., the probability distribution of the data. The hypothesis in a goodness-of-fit study in Chapter 4 is concerned with the univariate population distribution from which a set of independent variables is drawn. These hypotheses are so general that no analogous counterparts exist within the realm of parametric statistics. Thus these problems are more suitable to be viewed under nonparametric procedures. In a classical one-sample inference problem, the single-sample data are used to obtain information about some particular aspect of the population distribution, 157usually one or more of its parameters. Nonparametric techniques are useful here too, particularly when a location parameter is of interest.
