2.11 Define (Z) = E[tr ( Z)], for : p p, Z: p p, =...

2.11 Define ƒ(Z) = E[tr ( –Z)²], for : p × p, Z: p × p, = ’, Z = Z’, and the expectation is taken with respect to the distribution of . Show that ƒ(Z) is minimized for Z = E( ). It will be seen in (7.1.11) that Z is the Bayes estimator under a diffuse prior when ƒ(Z) is the loss function of interest. [Hint: See Section 2.14.1.]