ABSTRACT
Suppose that y $ \mathbf{y} $ https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781351264686/dcbe6876-28fa-432e-bea2-b6aed193917a/content/inline-math7_1.tif"/> is an N × 1 $ N \times 1 $ https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781351264686/dcbe6876-28fa-432e-bea2-b6aed193917a/content/inline-math7_2.tif"/> observable random vector that follows a G–M model. And suppose that we wish to make inferences about a parametric function of the form λ ′ β $ \lambda ^{\prime }\beta $ https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781351264686/dcbe6876-28fa-432e-bea2-b6aed193917a/content/inline-math7_3.tif"/> (where λ $ \lambda $ https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781351264686/dcbe6876-28fa-432e-bea2-b6aed193917a/content/inline-math7_4.tif"/> is a P × 1 $ P \times 1 $ https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781351264686/dcbe6876-28fa-432e-bea2-b6aed193917a/content/inline-math7_5.tif"/> vector of constants) or, more generally, about a vector of such parametric functions. Or suppose that we wish to make inferences about the realization of an unobservable random variable whose expected value is of the form λ ′ β $ \lambda ^{\prime }\beta $ https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781351264686/dcbe6876-28fa-432e-bea2-b6aed193917a/content/inline-math7_6.tif"/> or about the realization of a vector of such random variables. Inferences that take the form of point estimation or prediction were considered in Chapter 5. The present chapter is devoted to inferences that take the form of an interval or set of values. More specifically, it is devoted to confidence intervals and sets (and to the corresponding tests of hypotheses).
