4. Find the numerically largest Eigenvalue of 1 2 A= fl 3 0| and the corresponding Eigenvector. 2 0 -4 {A.U, M/J 2006, A/M 2008, M/y 2914 [LU NID 2014 NM, SNMIIA.U M/J 2007 SNM, A/M 2011 gq [AU M/J 2016 SNM RIS}AU N/D 2015, MII 2016 Rap (AU AM, 2018 R08 NMIIA.U_N/D 2019 Ry Solution: Note : To find the Eigenvalues solve 23 = 8,22 + 8,4 - S3 = 0... (1) where S, = Sum of the main diagonal elements = 25+3+(-4) = S; = Sum of the minors of the main diagonal elements fb °F afi | 13 = (-12~0) + (100 — 4) + (75 - 1) = -12-104+74 = 42 Use your calculator, | Working rule given in| this book || S3 = |A| = -308 . (1) @ a3 - 242? — 42 + 308 = 0 Solving, we get 4; = 25,1822, dy = 4.1379, Ay = 2.9558 So, the numerically largest Eigenvalue = 25,1822 Now, we will use power method 1 1%» ay Hom “li 40-)-- 12S) 8 (gs 1) (asamg 00444] = si (i 7 2a = 25.178 = 25%, AX, = AX= ned cue 2 0 4 anes = 25.1718% 0.06888,1 = 25.1826 | 0.0451 = 25.1826X5 0.0685 25.1826 =| 1.135 1.7248 25.1821 a (ad = | 11353 | = 25.1821 Jo0451| = 25.1821 x, Axs = | 7260 0.0685 We have reached the limit. 1 A= 25.1821 and the corresponding Eigenvector is | 0.0451 0.0685 5, Obtain by the power method, the dominant Eigenvalue and the corresponding Eigenvector, correct to two decimal places, for the matrix 2 2 2 % $4 94] taking [1 0 0)" as the initial approximation to this 1 % vector. [A.U. May, 1996] Solution : [AU M/J 2006 SNM, M/J 2007 SNM] Note : To find the Eigenvalues solve 3 — 8,2? + S)4—S3=0 ... (1) where S$, = Sum of the main diagonal elements Use your calculator, Working rule given in + aya 2 this book S =| Solving, we get 2, = 7.005, Ap = 0.5225, 43 = 1.6394 So, the numerically largest Eigenvafue = 72 2 2) |1 AX, = [2/3 5/3 5/3} |O] = |273| = 2 |1/3) = 2x, 1 5/2 11/2} |0 2 2 2 2 1 11/3) 13.67 0.80 AX)= |2/3 5/3 5/3] |1/3] = 137/18] = | 2.06 | =4.58 ie =4.58 x, 1 5/2 11/2] 1/2 55/12] |4.58 2 4 io [os 0.61 AX; = 2 5/3 5/3] 0.45] =|2.95] = 7.43 |0.40] = 7.43x, 5/2 11/2 1 743 1 2 2} fo61) [4.02 0.57 5/3 5/3) |0.40| = |2.74] = 7.11 |0.39] = TALXs s/2 11/2\| 1) {741 1 2 2 24\fos7] [3.92 0.56 2/3 5/3 5/3||039|=|2.70] = 7.05 }0.38] = 7.05 X, 1 5/2 11/2 1 7.05 1 vn AX, = = 2 2) [os6) [3.88 0.55 AX, = 5/3 5/3||0.38] = |3.01] =7.01 |0.43] = 7.01 X, 5/2 11/2 2 7.01 1 2 2) 10.55 3.96 0.56 2s. 5/3 5/3 a 2.75] = 7.13 |0.39 5/2 11/2 713 1 2 2 n 56 0.55 AXs = [3 53 53 039 20 = 704/038 5/2 11¢ 1 ae) 2 ass 0.55 XQ = 12/3 5/3 5/3 261 =7 es i 5/2 11/2 2 2 2] [oss 0.55 AX = |2/3 5/3 5/3/1038] = 7 |038 1° 52° uit" 4 1 Hence, the dominant Eigenvalue = 7 a) 1 HON Wen, Fo 0.55 Corresponding Eigenvector = |0.38 < primer erence11 ' Find the numerically largest Eigenvalue of A = |1 5 i by 6 3 lg power method. [AU AM 2019 (R-17)] Solution + ih tat X, = [1] be an arbitrary initial Eigenvector 1 bo1 3) it 5 0 AM= {1 5 4 ft) = 7} = 7/1] =x, Sh 4k 5 0.7 ele 3/027) 38 0. AXM= |1 5 1) ] 1) = [64] = 64/1] = 64x, 3 1 1} +07, 38 0.6 1 1 3) fo 0. AXs= |1 5 af [a 62]}1] = 62x, Ege a bs fir) 06 Here, X3=X4, Hence, the numerically largest Eigenvalue is 6. I Poe ABR W SHORT QUESTIONS AND ANSWERS 1 Find the power method, the largest Eigenvalue of i a correct to two decimal places, choose [1, 1]7 as the initial Eigenvector. TAU, 1996] Solution ; a ff ton: Let X; = [i] MG = [4 1) fa] _ (s] _ 1 = 5X, EG E--s] = AXpm /4. af [4 48] _ 1 = 48%: : [ 4 [oa a [s4] es liz : ax [4 af a] _ fan) | = ATX, J [ 4 ba 7 al rae [oc] 467 Xs " oa pees, os 3 as w Pa 2a eee - a = 3 ryAX; = i 3] fe = [255] = 4.65 lo 6| = 45x, a * iy H| FA ms [ss] = [aal} = 463X, aXe r | [ot = (es etd [oa] = 462X, AXg = i 1 [ni = (:2] = 4a [aga = 462X, The Eigenvalue = 4.62 and Th ding Eigenvector = |, 2 e corresponding Eigenvector = |q 69 2, Determine the largest Eigenvalue and the corresponding Eigenvector of the matrix i 1) correct to two decimal places using power metho Solution : [A.U. Oct., 1996, CBT M/J 2010] [A.U N/D 2019 R-17) 1 1] ji 2 ef Wf) G w= ELE) -2f) os ~. This shows that the largest Eigenvalue = 2 = 2X) i} n =e ‘The corresponding Eigenvector = (i 3, The power method will work satisfactorily only if A has an .. Eigenvalue. Solution : Dominant 4, If the Eigenvalues of A are -3, 3, 1 then the dominant Eigenvalue of # is .. Solution : No dominant Eigenvalue 5. If the Eigenvalues of A are 1, 3, 4, then the dominant Eigenvalue “* is Solution : 4.er ——— igenvalues of A are —4, 3, 4, the Eigen 1 then the dominant Eigenvalue of A : - 4 solution = How will you find the smallest Eigenvalue of a (or) Write down the procedure to find the numeric; of a matrix by power method. Square matrix A? ‘ally smallest Eigenvalue [AU AM 2010) By power method, the largest Eigenvalue of Aq! can be found. Then, smallest Eigenvalue of A is the reciprocal of largest Eigenvalue of Aq}, 1 . Explain power method to find the dominant Eigenvalue of a square matrix A. (or) Define power method. [A.U A/M 2019 R-13] If Xp is an initial arbitrary vector, then compute Yet Yuet = Ax, and X41 = Meat where M,41 is the numerically largest value of Y,41- i : [Yea | Then dominant Eigenvalue 4, = tim PR; yy koo [Yili 9. When do we use the power method? [A.U A/M 2015 (R8)] Solution : We use, the power method to find the largest Eigenvalue in magnitude and the corresponding Eigenvector of A. 10. When can we expect faster convergence in power method? Solution ; We assume, |dy | > [Ap | >... > Mal. Faster convergence is obtained when |Aq| < | 41| IL. Does the Power method give the sign of the largest Eigenvalue? Solution ; No, [A.U M/J 2013] R, What is the use of power method ? Solution ; To find the numerically largest Eigenvalue of a given matrix.