Finding Inverses by Elementary Row Operations. ‘The method for finding the inverse of a nonsingular matrix by elementary row operations is analogous to the method we used for solving systems of linear equations. In fact it looks like solving the system of equations A-@ =. Inevery step we clear a whole column, ie. the column exists only of zeros except for the cleaning element.Example | 1 02 Find the inverse of A=|2 —1 3. 4°18 First, write down the 3x3 matrix whose first 3 columns are the columns of A and then, after a vertical line of dashes, the 3 columns of the 33 identity matrix. Then choose at the left side of the dashes nonzero cleaning elements to clear a whole column in one elimination round. If the matrix is nonsingular, we can choose out of all of the three rows one nonzero cleaning element. 1 02|100 @ oo 2;100 fj 0 2) 100) 2-13/010 6/6 O11 | 01 1 #/]0 0 -1 | 611 4() 8/001 4 fl) 8/001 of) 8 |-401 f] 0 2) 10 0 fi] 0 o| -1 2 2 0 0) | 6-1-1) + ]/0 0 ff} | 6 -1 - oe of] o|-4 0 1 of] o| -4 0 1 100-11 2 2 o10 | -4 0 1 ooi1) 6 -1 -1 At the left-hand side of the dashes stands the 33 identity matrix, on the right-hand side the inverse of A. (-11 2 2) SoAt=]-4 0 1 (6 -1 -1) Example 2 123 Find the inverse of S=|2 5 7 3.7 10 M23 |100 fi) 23 -2 0| 257 |010¢]0 @1 10 3 7) 001 ot “11 The process of inverting the matrix has stopped prematurely. During the process we encountered left of the dashes a row that completely exists of zeros. In that case the matrix has no inverse matrix and is called singular.Example 3 3 Find the inverse of A =|2 1 32 21 1 () os on) Example 4 Solve the system: 3a, +22, + 5a, =1 2n, +2, 3 The coeficients of the system are exactly the entries of matrix A ofthe preceding example, so we ean write: 1 A-# =D where | =|2| and so, by theorem 16.6.2:‘When we write it out on a scalar way Cramer's Rule looks like: br, = 4° =P the otters a,b, 6 d,p and q are given real constants. cx, +2, =4 ‘The solution is Of course under the condition: ad — cb 0 Definition: ‘The minor of an element a,, from a (3,3)-matrix A is the determinant of the (2,2)-matrix that comes from A by deleting row i and column. 15 The mir ft, of Ate ati 3 6/7 iXO-8x(-8) = 21Definition the cofactor of a, by ¢ Example: The cofactor of a, in the matrix from example 5 is: ¢ -1)x21=-21 Example 2 3-1 1 4 5 |=2-(-17 +3-(-1)'- 7 PR PEN Tg g/t a 3 6 3.8 6 2-(41)- (24 — 40) +3 -(—1)-(6 +15) + (-1)-(+1)-(8 +12) For the calculation of the determinant we have chosen the first row. If' we choose the second column we find: 23-1 1 4 5/=3 3.8 6 —3-(6 +15) + 4-(12—3)—8-(10 +1) = ~63 + 36-88 = -115 This is named “expansion of the determinant along the second column”.Example 14 3/=(-3)-(-9} 6-(-1 =3).(= oy. -(-1)- ; -5 6 0 Here we have an expansion along the second column, Expansion along the third column yields: —3x(-3)+0=9 3x15 +4x0-6x6=9Then: Example 3 Let A=|2 - Verify that |A|=7. Now we write down the matrix of cofactors: 1 4 5 1 3-2 =I |. The transpose gives the adjoint matrix: -11 19 -1 4 3 -1 4 3 -1l adj(A)=|-S -2 19 | so the inverse is: ated -5 -2 19 1-1 -l 1-1 -103 02) Let B= Then | B|= 0.30.7 -(—0.4)x(—0.2) = 0.21-0.08= 0.13 et ( 07 hen | B| = 0.30.7 —(—0.4)x(-0.2) 1 _. 0.7 04 0.7 0.2 the matrix of cofactors is so adj(A)= 0.2 0.3 0.4 0. 0.7 0.2) We find that Bo! = 1 ( 0.13(0.4 0.3)