Question: 5.22 This exercise is regarding Example 5.8 (continued) in Section 5.3.2. For parts (i) and (ii) you may use the following matrix identity in Appendix

5.22 This exercise is regarding Example 5.8 (continued) in Section 5.3.2. For parts

(i) and (ii) you may use the following matrix identity in Appendix A.1.2: (D±

BA

−1B



)

−1 = D

−1 ∓ D

−1B(A ± B



D

−1B)

−1B



D

−1.

(i) Define H = δI + Z



Z. Show that B(γ ) = δ(X



X − X



ZH

−1Z



X)

−1X



(δI + ZZ



)

−1.

(ii) Furthermore, let Q = δI + Z



PZ. Show by continuing with (i) that B(γ ) = (X



X)

−1X

{I + ZQ

−1Z



(I − P)}(I − ZH

−1Z



).

(iii) Continuing with (ii), show that B(γ ) = (X



X)

−1X



(I − ZQ

−1Z



P).

(iv) Show that (X



X)

−1X

 = λ

−1/2 min (X



X).

(v) Write A = P



Z. Show that the positive eigenvalues of S = A(δI +

AA



)

−2A

 are λi(δ + λi )

−2, 1 ≤ i ≤ m, where λi, 1 ≤ i ≤ m, are the positive eigenvalues of AA

. [Hint: The positive eigenvalues of S are the same as those of U = (δI + AA



)

−1A



A(δI + AA



)

−1.] Use this result to show

(δI + A



A)

−1A

 ≤

Z

min λi > 0

λi

, δ>0, where λ1, . . . , λm are the eigenvalues of A



A.

Step by Step Solution

There are 3 Steps involved in it

1 Expert Approved Answer
Step: 1 Unlock blur-text-image
Question Has Been Solved by an Expert!

Get step-by-step solutions from verified subject matter experts

Step: 2 Unlock
Step: 3 Unlock

Students Have Also Explored These Related Principles Of Managerial Statistics Questions!