When the Distribution Is Known

Abstract

One of the strengths of the hypothesis-testing procedures described in the preceding chapter is that you don’t need to know anything about the underlying population(s) to apply them. But suppose you do know something about these populations, that you have full knowledge of the underlying processes that led to the observations in your sample(s), should you still use the same statistical tests? The answer is no,not always, particularly with very small or very large amounts of data. In this chapter, we’ll consider several parametric approximations, including the binomial (which you were introduced to in Chapter 2), the Poisson, and the normal or Gaussian, along with several distributions derived from the latter that are of value in testing location and dispersion.