HELON FORM OF A MATRIX holon form if it has the following three properties + entry in the row should be 442 ROW EC! x is said to be in row Definition. A matt 1, arowdoes not consist entirely of 2¢ i, called the leading 1. »ros, then the first nonzero entirely of zeros, then the leading | in the the leading Ifany two successive TOWS that do not consist Jower row occurs farther fo the right than the leading 1 in the upper row, 1.e., } appears farther fo the right as we move down the rows of the matrix.™ Business Nagy ne my 7D : that consist entirely of zeros, then they are gronpeg Wty are any rows that consis Siena Example 36. The following matrices ; i> 0 oj] ff 0 ie O07 8 jo 10] a - 0001 000 Ae it ow is I, the leading | . re lon form, because the first nonzero entry in each re i att a “Phe ea down the rows, and any rows that consist entirely o Zet05 jove farther to the right as we m i at the bottom of the matrix, ' oa reasong : le 37. The following matrices are not in the row echelon for Stay xample 37, in the bracket : iS. 2 : : 0 8) (The leading 1 in the last row is not to the righ of 070 120 the leading 1 of the middle row) 020 ; i 001 (The leading entry of the first row is not 1) 000 5 5 (The row of zeros is N01 at the bottom) The step-by-step procedure sform any matrix f0 row echelon form is outlined as follows: Step 1. Locate the lefimost column that does Step 2. Interchange the top row with another ro 1¢P Of the column found in Step 1. Sep 3. the entry that is now at the top of the colun row by 'a"in order to introduce leading 1, Step 4. Add suitable muttipte &f the top row t0 the found in Step 1 ig ee ivice the fiat rows below so th : 80 that i leading I become zeros, all entries below the Step $. Now cover the top row of the matrix anet 9 submatrix thar is left. Continue in this way unit thes e matrix 5, (PPlied to the ci ‘ form, ‘S12 row echelon The procedure is best illustrated with the help of followin,Matrices and Determinants fo toa -4 2 4 4 oe 7 i to row echelon form by using 38, Reducethematrix, © | Exampl row operations, Locate the lefimost colin clemet an that does not consist entirely of Zeros Step I. Solution o laa 2 4.8 6 2-40 1 1 eftmost nonzero column 2 I Step 2. Apply R, @ R, fo get_a nonzero entry at the top of the first column found in Step I 248 6 0 1 3-4 [2-4 0 1 1 Step 3. Apply WZ R, in order to introduce a leading 1 fl 2 4 3 0 13-4 [2-4 0 1 so that all entries below the leading 1 become zeros. step 4. Apply Rs > Bs — 2 fi 2 4 3 o 1 3-4 [o -8 -8 -5 ix and start again with Step 1 applied to the submatrix Step 5. Nowcover the top row of ‘the matri that is left. 1 2 4 3 o 1 3-4 0 -8 -8°-5 4 — Lefimost nonzero column inthe submatrix Applying Ry > Rs + 8a to introduce a zero below the leading |.Pe, | Business ny 5 “hem 152 m [2 428 o 1 3-4 0 0 16 -37 TD Lefimnt nonzero column inthe new submunryy 1 a Applying Ry > 16 Ry to introduce a leading 1. E23 013 -4 ee if 16 The last matrix is in row echelon form. i234 Example 39. Reduce the matrix A=|2 -1 -2 0/10 row echelon form bY Using 1-3-3 - elementary row operations, Solution. The given matrix is : 23d A=|2 | =2 9 Lesa 9 Applying the operations Ry —> Ry ~2R, and Ry —> Rs R, to get fo 0 -5 -3 -g 0-5-6 6 I Applying the operation Ry > Re to get ty oy 8 8 a: ts 0 -5 -6 -6 Applying the operation Ry > 2, + 5k, to get Los Gg 0 I 2 2 SS Oywrorminants and DE / 15: ne operation Ry xh to get : wi a Aenan ol g 00 1 ig in FO" echelon form. ht i320 6 40. Reducethematrix A = pg sores poor 0 0 5 10 15| t° fOW echelon form b: element tary TOW operations. 26 0 818 see given matrix is ¢ 113 -2 0: 0 SG ss A= j0 0 5 10 15 Ibo 0 8 18 ait operations Ry > Ry ~ 2K, and Ry -> Ry ~2R, toget 5 9 8 0 0-1 -2 -3 0 0 5 10 15 fo 0.34 8 Applying the operation R, > (-1) Ry to get ri 3%—2-0 0 oo & 2°'3 ov or, 3 10 15 OrOea . 18 Applying the operations R, > Rs ~ 5R, and Ry > Ry -4R 19 eet a 2 00 0.0. hed. 3 oo 0 0 loo 0 0 6 Ny ng the operation Ry <> Ry to getNay My U7 ‘ sky to pel Ria Sha ing the operation ROR Applying the eee ot 6 4 00 4 -M] 1 ret —R, to ge ion Ry >= ks Applying the operation Ry > 3 {2 <1 6 o1 6-4 00 1-1 cl rm, we apply the operaij helon form. To find the reduced echelon form, pply PCration which is in row echel . RR, +Ryand Ry Ry -6R; to get 1205 0102 001-1 Applying the operation R, > R ~ 2R to get A Yas Vd ee 0102 00 1-1 The last matrix is in reduced row echelon form 3.2 a 1 Example 44, Reducethematrix A = | 2 3 0-1 |t0 reduced ae €chelon f . Tao 347 form by using elementary row operations. ‘Solution. Applying the operation ROR, to get [ pO 3-7, 23 0-1 Ba 2 at Applying the operations Fz > Ry 2K and Ry > Ry -3h, 1-6 3 7 0 15 ~6 ~15 0 20 -8 -2 to getp Determinants vs and Pa 197 eration Ry = Ry to ge ane operation 15 “210 ket nn AP 1-6 3 9 Oe = -l 0 20-8 ~29 rine bs opsration Ry 9 Ry — 20R to yet apne n 1-6 3 7 \ 2 oO 1-24 5 0 0 0 9 wich i 9 row echelon form. To find the reduced row echelon form, we a eects We apply the operation = 1 0 bg 3 1 ore a 0 0 0 0 ‘he last matrix is in reduced row echelon form. Remark. It may be remarked that every matrix has a unique reduced row echelon form; that . one will get the same reduced row echelon form for a given matrix no matter how the row operations are varied. 413. RANK OF A MATRIX Rank of a matrix is one of the most important concepts in matrix theory. It plays an important role in the study of solutions of a system of linear equations. Definition. (r- rowed minor of a matrix) Let A be an m * n matrix and r be a natural number such that r < min (m,n), where min (m,1) denotes the smaller of m and 7. Then the determinant of a x + square submatrix of A obtained on deleting any (m — r) rows and (7~r) columns of A is called an r-rowed minor of A. Note. An r-rowed minor of a matrix A is also called minor of A of order r. Definition. (Rank of a Matrix) A non-zero matrix A is said to have rank r if at least one of ‘is rowed minors is different from zero while every (r + 1)-rowed minor, if any, is zero. A “ro matrix is said to have rank zero. Notatic Puation. The rank of a matrix A is denoted by p(A).Mh ng 18k nih ‘Many, Remarks. ca Wo The rank of a non-zero matrix must be at teas A) min Gn, ny 2. Wt isa matrix of order mm, then pC) s min ( 3% Weve ix 4 ip x0 PHP A does NOL ponieny “rowed minor of a matrix is Zora (or IEA docs HOt poweny yy , minor), then p(t) least one y-rowed minor of a matin A whieh ie different jy, a: n IPA is a non singular matrix (ic. [Al # 0) of order n, then p(A) = », the rank of an identity matrix of order 1 is 1. 6. The rank of a 7. The elementary row (or column) transformations do not alter the rank of a Matriy Example 45. Find the rank of cach of W pritieg, ris A is same as that of its transpose, Ae, P(A) = p(A%) the following matrices : : 123 io i) a jw {2 1-1 a3 : 2-11 ens hari 12-1 4 i) | OO") pea ne Unix, 2011 Oe tes 7 “1-2 6 -7 Teele Selution (i) Let A=12 1-11. since 4 ig Square matrix of order 3, therefore 2-1 1 : PLA) < 3. Now IAI = 10-1) -1(2 42) *1-2-2) = 04g 8x0 =-8% 4 is non-singular and hence pA) = 3, dans () Leta =13 4 51]. Since a is square matrix of org a Of order 3, therefore pcg) < 3. Now |A[ = 1G2 ~ 30) ~2 (24 - 20) + 3.18 — 16) = hae Set 6 tg= and Determinants wromices an : inor of A of order 3 is zero, Hence Tas ao enact zero, Hence p(A) <3. However, there is a non-zero of Ao! order 2. ino © 12 \; ‘- 20 play = 2 ol an we el 0 1. Since A is a square matrix of order 3, therefore, (A) < 3. Now rol moe 4] = 0 (2 all the elements in the second column are zero) sus ine only minor of A of order 3 is zero, Hence {) < 3. The minors of A of order 2 are: 10 0 a 1 11 10 oif> fia in fact it can be easily seen that all the minors of 4 of order 2 are zero. Thus pa) <2. Frowever, the matrix A is non-zero, therefore p(A) = 1. 0, etc., 1 2-1 4 (iv) Let A=|2 4 3 5 -1-2 6 -7 since matrix A is of order 3 x 4, therefore p(A) < 3. Let us consider minors of order 3 = tel 1-1 4 12 4 2-1 4 M,=|2 4 3)Ma=|2 3 5|,M;=\2 4 S\andM,=|4 3 5 -1 -2 6 -1 6.-7 1-2 -7 2 6 -7| Now, M, = 1(24+6)-2(12 +3)-1-4 +4) =0 Similarly, it can be checked that M, = Ms = Mg = 0. Thus, all minors of A of order 3 are zero, therefore p(A) < 3.Now we shall see minors of order 2. We have “1 le |-s+a-1040 4 p(A) = Example 46. Find the rank of each of the following matrices + eal eeel etl 5 3 14 4 ‘ : Gio oe [BBA LP. Univ. 2011 ie 2031 t Sais iHOLHSBusiness ines Ay lathe, 160 = 4 See ‘ 1]. since mnuin A is oF onder 34, theretore pa, a 0 1 Now Araneae oar Let us consider the following minor of or a = 5(2+2)-0+1(6-14) = 20-8 = 12 44 M=/0 1 2 oe ifferent from z mae per Thus there exists at least one 3-rowed-minor of A which is diffe: CTO. He pA) = 3. Dee U6 Leti ad m2 21 Sinead ialatecuare aiervoF onder 4ythenaros pays 20-31 3930 3 3-2 1 A= 2h aS 13 - l4=1fo -3 1p -3 V+1/2 0 1-1/2 0 -3| Now 3 0 3 3 0 3 3 3 3 SS 8A) = 1-24) -1(6) + (- 6) 1-30) = ~24-6-64+30=-6 40 Thus the only minor of 4 of order 4 is non-zero, Hence p(A) = 4, Finding the Rank by Transforming the Matrix into Row Echelon Form We now state a theorem (without proof) which can be used to find the rank of a matrix b first transforming it into row-echelon form using elementary Tow operations, : Theorem 4.2. The rank of a matrix is the fmunber of nonzero rows in i i een transformed to row-echelon fori using elementary row operations, ™t® Hier it has IS. ee Example 47. Find the rank of the matrix A=/ 2 4 5 ipa by ‘ransforming it into row-echelon form, Solution. Applying the operations a > Ry -2R, and Ry 5 R+ Rt 110 ger Te rg OF Osa eg eee 5 -3‘ Vp lol ynntion Ry > 5 Arto pot re © a ie 4 | i 0 0! s| i eo 3 L sation Ry > Ry = SRy to get the oper" 2 vite : rl 2 -) 4 | 3 0 1-= : 5 jo 0 0 Oo ae matrix is in row-echelon form having two nonzero rows, therefore the oa e ay giants is 2 aa} 5 27} 3 7. by transforming it into 4 =3 1 sample 18. Find the rank of the matrix 4 = co RN echelon form. 1 Suution, Applying the operation 2, > 3h to get protire 3. 5 B 8289 ae eG 844 7 e813 8 4.55 {| ‘hing the operations Ry > Ry —4R,, Ry > Ry —8R, and Ry > Ry —8Rjto get i 3 7) 2 Co 7 esBusiness May, 162 5 aD foe | eae ' 7 5 0 oe! 0 0 21) 4 5Ry and Ry > Rs -15 4 15R, to get Applying the operations Ry — Ry fi fa a 8 7 co 1 0 0 lo 0.7010 The last matrix is in row-echelon form having two nonzero rows, therefore the rank of iy original matrix is 2. : : Example 49, Find rank of the following matrix using both determinant and using echelon fom, 364 re 912 3 [BBA LP. Univ. 2005, Solution. Finding rank using determinants: Since matrix A is of order 3 * 4, therefore rank (4) < 3, Let us consider minors of order : 13 4 ta3 13h 4 M,=|3 9 12) M,=|3 12 3, M,=|/3 9 31 a, =|o 12 3 La ea ce 134 1 j13 4 [a No [3 9 12/=3/3 3 12)=3x0= [ia fea Similarly it cam be checked that Mf, = MM, = My = 0. Thus a minors of 2010, therefore rank (A) <3. Now we shall examine minors op ondat 3 ee toot order 3 a : er 2. We have 43 12 3)= 72449 + mk (A) = 2,y we wa ann wie Eehelon Form: a ane operations RR -3R ond a, Ria g : « a Ryd pet | 4 , 900 9 900 4 sg tte onerion R, = R, to get yf Hietatieg ce 000 + 000 9 sing the operation Rs -5 Ry to get 1 34 3 000 0000 tekst matrix is in now-echelon form having two nonzero rows, therefore the rank of thi cool matrix is 2. ; EXERCISE 4.3 -1 0 yar eer) Lien A=|2 4 5| and B=/5 1 7], verifythat 6 23.8 (i) (A+B) =A'+BY, (ii) (ABY'=B'A', (ii) (QAY'=2A8" 1 5 $3 * Findtheadjoint of the matrix A= | 5 a A 1 = (adj VAY hi H= (adi rthe square matrix A= | — Ih one 140 1 2 |, verily that A (adj A)= | A 002 [BBA LP. Univ 2010) a ind the; dhe inverse of the matrix A= eve 5