From now on we will assume that only \(\hat {\theta }\) and \( \mathop {\mathrm {Var}}\{\hat {\theta } \}\) or some functions of these quantities are used to describe the results of an experiment. This is justified by the fact that in many cases and in particular in the case of normally distributed observations, \(\hat {\theta }\) and \( \mathop {\mathrm {Var}} (\hat {\theta })\) contain in some sense all information that is available from an experiment with respect to the linear model \(y = \theta ^T f (x) + \varepsilon \) [cf. Rao. Linear statistical inference and its applications (2nd ed.). Wiley, 1973, Chpt. 2d].