Matrices & Determinant iS j Matrices & Determinant 2s for everything else, 90 for amatharatical theory, beauty can be parcelvedibut not explained = Cayley Artur introduction : Any rectangular arrangement of numbers (real or complex) (or of real valued or complex valued expressions) is called a matrix. if a matrix has m rows and n columns then the order of matrix is written as m * n and we call it as order m byn The general m * n matrix is a ay ats ay an Aa 2 Aas 89) an ay a2 ap ay Amt @me Ana Bay Benn where a, denote the element of i row & j* column. The above matrix is usually denoted as [a)],,., Notes: () The elements a,,, aj, ayy... are called as diagonal elements. Their sum is called as trace of A denoted as tr(A) (i) Capital letters of English alphabets are used to denote matrices. (ii) Order of a matrix : a matrix has m rows and n columns, then we say that its order is "m by written as "m xn" Example # 1: Consiruct a 8 x2 matrix whose elements are given by a, = +1i-3)| a1 a2] Solution: In general a3 2 matrixis given byA= | a2) az2 831 sz | 4 3 2, 3and 2 aj= Gli 3121, 2,3.andj=1, 1 1 5 Therefore a= S19 x1 ]=4 aye Fi1-3x2i= 5 a1 a1 4 . aye Zl2-3x11=5 ag =F 12-3x2/=2 1 1 3 ay,= 5 13-3%11=0 ay = G19-3x21=5 15 2 1 Hence the required matrix is given hyA=|z 2 03 2 ‘Types of Matrices : Row matrix : ‘A matrix having only one row is called as row matrix (or row vector) General form of row matrix ISA [Bye, A491 M36 on Aya) This is a matrix of order 1x n" (or a row matrix of order n)Matrices & Determinant @ Column matrix : ‘A matrix having only one column is called as column matrix (or column vector). ayy aa Column matrix is in the form A= Bn This is a matrix of order "m x 1° (or a column matrix of order m) Zero matrix : A= [ajJq x (8 called a zero matrix, if a, = 0 Vi 8] 00 0} yooo ea:0[o oo! (fo 00 lo oo Square matrix : ‘A matrix in which number of rows & columns are equal is called a square matrix. The general form of a square matrix is, an ae an an an Bon A which we denote as A= [aj). Be Baz ve Ban This is a matrix of order "n xn" (or a square matrix of order n) Diagonal matrix : A square matrix [a)], is said to be a diagonal matrix if a, = Ofori z. (Le., all he elements of the square matrix other than diagonal elements are zero) Note : Diagonal matrix of order nis denoted as Diag (ayq, yy. gq) : aooo aco oboo eg.:()|? > % qfo ooo doc} oo0c Scalar matrix : Scalar matrix is a diagonal matrix in which all the scalar matrix, if (i) aj = 0 for i j and (ii) a= k for i .gonal elements are same, A= [aj], is a aoo ao oad eg. ( | (i) oa looa Unit matrix (Identity matrix) : Unit matrix is a diagonal matrix in which all the diagonal elements are unity. Unit matrix of order 'n' is denoted by 1, (orl) ot oo1Matrices & Determinant @ Upper triangular matrix : ‘A= [ajlq xq i8 Said to be upper triangular, i a, = 0 for i > j (i.e, all the elements below the diagonal elements are zero). abcd abe eg.) Ox yz Moxy oouv ooz Lower trlangular matrix : ‘A= [2m en i Said to be a lower triangular matrix, ifa, = 0 for i A+B e My. fF) (b) A+B=B+A () (A+B) +C=A+(B+C) (d) = [0], ., is the additive identity. (@)_ ForeveryAc M,,. «(F), —Ais the additive inverse. © (A+B) =2A+98 (g) (h) 8 0 Example #4: IF A=|4 ~2/ andB “then find the matrix X, such that 2A + 3X = 6B Solution: We have 2A+ 3X = 5B. > 3X=5B-2A 1 > 3 (6B-2A) af{ 2-2) [8 oy ff -101 = 5 4 2[-44 -2[]= 3] 20 | L[-s 1 [3 6 \|-25 5 | {[19-16 -10+0) {6 -10 = z| 20-8 10+4 | 4 4 -25-6 s-12} [31 7] |-31Matrices & Determinant ® Multiplication of matrices : Notes LetAand B be two matrices such that the number of columns of Ais same as number of rows Of BLA [ilney & B= [Dil Then AB = [cjJq., where 6; = >) 8xby, which is the dot product of i row vector of A and jt column vector of B. 123) 34917] 234] 1372) eg.:A |B :(1) The product AB is defined iff the number of columns of Ais equal to the number of rows of B.Ais called as premultiplier & B is called as post multiplier. AB is defined =& BAis defined, (2) Ingeneral AB # BA, even when both the products are defined (3) A(BC) = (AB) C, whenever itis defined. Properties of matrix multiplication : Notes: Example #5: Solution : Consider all square matrices of order’. Let M, (F) denote the set of all square matrices of order n, (where Fis Q, Ror C). Then (@) ABeM,(F)>ABe M, (F) (b) —_Ingeneral AB * BA (c) (AB) C=A(BC) (4) I, the identity matrix of order n, is the multiplicative identity. AL,=A=1,A VAcM, (F) (e) For every non singular matrix A (L.e., [A| + 0) of M, (F) there exist a unique (particular) matrix B < M, (F) so that AB = 1, = BA. In this case we say that A& B are multiplicative inverse of one another. In notations, we write B= A" or A= B-! () If his a scalar (2A) B = (AB) = AB). (@) A(B+C)=AB+AC VA.B,C eM, (F) (n) (A+B)C=AC+BC VA.B,CeM,(F) (1) LotA= [aj x» Then Al, =A &1,,A=A, where I, & 1, are identity matrices of order n &m respectively, (2) Fora square matrix A, A? denotes AA, A? denotes AAA etc. 123 was | 72 4) th x = =|) j [then show that A® - 230-401 = 0 1 2 3]/1 2 3] [19 4 8) Wehavear=aa=|2 -2 1] |3 -2 4 |: 2 | 42 1}|4 2 1] [14 6 15] 1 2 3) [19 4 8) [63 46 69 so atcan=|3 -2 1]]1 12 8] |e9 -6 23 4 2 1} |14 6 15] |92 46 63 fe3 46 69] [1 2 3] [10 0} 92 46 63 4214 oo4 dakshnanaMatrices & Determinant @ 63 46 69] [-23 -46 -69] [-49 0 0] = |69 -6 23] ,|-69 46 -23),] 0 -40 0 92 46 63] |-92 -46 —23 0 0 -4o} 63-23-40 46-46+0 69-69+0 69-6940 -6446-40 23-2340 |90-92+0 46-46+0 63-23-40 000 =|0 00/9 ooo Self practice problems : cos -sino _ (1) FA =T Sine cose | Very that Ala) A(B) =Ala * 8) Hence show that in this case A(a). A(B) = A(B) . A(a). 4 6 - 2 4) 2) Leta=]3 0 2| B=|0 t/andc=(3 12) 1-25 12 Then which of the products ABC, ACB, BAC, BCA, CAB, CBAare defined. Calculate the product whichever is defined. Answer (2) Only CABis defined. CAB = [25 100] Transpose of a matrix : Let A =[a\]q.x- Then the transpose of Ais denoted by A’( or A") and is defined as A’= [bjlqam Where by =a, Vi8) i.e. A’ is obtained by rewriting all the rows of A as columns (or by rewriting all the columns of A as rows). - ; tax 234 aby eg. :A=[2 Po a | Bez Ixy zw 4dw Results: (j)_—_-For any matrix A= [a,)qs y (A’ (i) Let bea scalar & Abe a matrix. Then (2A)' = 2A" Gi) (+8) A’ +BY & (A- B)' =A’ — B' for two comparable matrices A and B. WW) (Ap EAS ca BAY SAY AZ toes t Ay! where A, are comparable. "AY — LetA= (a), x, &B= [bi], (Vi) (AG Ag cess)! eas then (AB) = Ay! enn! Ay', provided the product is defined Symmetric & skew-symmetric matrlx : A square matrix Ais said to be symmetric if A’ = A ie. LetA=aj),. Ais symmetric if a; = a, Vi &j. Asquare matrix A is said to be skew-symmetric if A’ = —A ie, LetA= aj, Ais skew-symmetric iff a, = — a, Vi &j.Matrices & Determinant @ 9 eg A | is a symmetric matrix. © fo x y x 0 Z| is a skew-symmetric matrix. y -z2 0 Notes: (1) Ina skew-symmetric matrix all the diagonal elements are zero. (:a=-a, > a=0) (2) For any square matrix A, A + A’ is symmetric & A—A’ is skew-symmetric. (3) Every square matrix can be uniqualy expressed as a sum of two square matrices of which one is symmetric and the other is skew-symmetric. A=8+c,wherea= 1 (Asay aC = 2 Example #6: IfA : .B=[1 3-6}, verify that (AB) Solution: — Wehave B=[1 3-6] [-2] -2 -6 12 si -24 5 15 _30 ; ‘| 4 a4 6 abs 2 1) -csy 12 -24 -30 2-2-4 Example #7 : Express the matic =| 1 24) as mo eum oe smmatre anda stow amet matric aa 4 Solution Here 3-2Matrices & Determinant @ Example #8 : Solution Now Thus o 1 Also, Let z 3 - [2-3 0 2 2 Thus, B is reresented as the sum of a symmetric and a skew symmetric matrix. Show that BAB’ is symmetric or skew-symmetric according as A is symmetric or skow= symmetric (where B is any square matrix whose order is same as that of A). Case-1Ais symmetric = =A (BAB’)’ = (B')'A’B' = BAB’ > BAB’ is symmetric. cas Ais skew-symmetric => W=-A (BABY = (BAB =B(—A)B" (BAB") = BAB" is skew-symmetric Self practice problems @) (4) Submatrlx : For any square matrix A, show that A’A & AA’ are symmetric matrices. If A & B are symmetric matrices of same order, then show that AB + BA is symmetric and AB - BA is skew-symmetric. Let be a given matrix. The matrix obtained by deleting some rows or columns of Ais called as submatrix of A. abe dl eg. xyzw par sl] ac abe abd Then fx 2]. [0 9 gl) [x ¥ 2] areall submatrices of A P ~ IparMatrices & Determinant @ Determinant of a square matrix : To every square matrix A = [a] of order n, we can associate a number (real or complex) called determinant of the square matrix. Let A= [a],,, be a 1x1 matrix. Determinant Ais defined as [AI eg. ESI IAL=-3 a ‘a bi Let A= , then |A| is defined as ad — be. cd 5 3 eg. A=|_y 4|-IAl=23 Minors & Cofactors : Let A be a determinant. Then minor of element ay, denoted by M,, is defined as the determinant of the submatrix obtained by deleting i* row & " column of A, Cofactor of element a, denoted by C,, is defined as C, = (- 1)'*/M, eg.4 eg. 2 Determinant of any order : Let A= [aj], be a square matrix (n> 1). Determinant of Ais defined as the ‘sum of products of elements of any one row (or any one column) with corresponding cofactors. ay Aye Ars egt A=]82 a2 25 a1 Ase Ass IAD ay Cy, # ay Ca + ajgC yg (using first row). a Ae ay) Ase an Ars aa Aas As Ass IAL = ay2 Cyz + age Cop + aggC3e (using second column) ay ays 81 2s = ay, 82 lar ass Aor Aes ayy Ata Aa Aas As Aas Transpose of a determinant : The transpose of a determinantis the determinant of transpose of the corresponding matrix la, by cy lay a, ay D=|a, b c,| => D'= fy b, by las bs cy ky ¢2 cyMatric es & Determinant ® Propertles of determinant (1) ie. eg. 8) (4) 6) 6) Mm (8) (9) (10) (1) IAI=[A’[ for any square matrix A. the value of a determinant remains unaltered, if the rows & columns are inter changed, ar by ca fay ay asl by C2] = |b; be bs| a bs Cs] fey cz ca] If any two rows (or columns) of a determinant be interchanged, the value of determinant is changed in sign only. lar by | 2 be Go| Let D,=[@2 bz | & p,=|@ bi &| Then D, las bs | a Bs 9 Let d be a scalar. Than 2, |A|is obtained by multiplying any one row (or any one column) of [AL by 2. lay by cy ka, Kb, Ke, D=|a2 b, ¢,| and E=|@ bz, Then E=KD jas bs sl a3 by Cs [AB pal AILBI. an A], when A. (aj), A skew-symmetric matrix of odd order has deteminant value zero. Ifa determinant has all the elements zero in any row or column, then its value is zero, an) ie D= 2 bz & 5 bs Cy Ifa determinant has any two rows (or columns) identical (or proportional), then its value is zero, la, b,c ie. la, by & 3 bs Go| If each element of any row (or column) can be expressed as a sum of two terms then the determinant can be expressed as the sum of two determinants, i.e. larix bry eyt2] fa, by o| |x yz a bp 2 |=|a2 be c2|+ fz by cp] ay bs 3] fas bs cs] fas bs 69) The value of a determinant is not altered by adding to the elements of any row (or column) a constant multiple ofthe corresponding elements of any other row (or column), la, b, | lay+may b,+mb, cy +mo,| 2 be 62] and D,=| a by 2 |. Then D, la bs cy] lag +na, by +nby cy +ne, Let A = [aj},. The sum of the products of elements of any row with corresponding cofactors of any other row is zero, (Similarly the sum of the products of elements of any column with corresponding cofactors of any other column is zero)Matrices & Determinant @ abe Example #9 Simplify |b c a cab Solution: Let R,>R, +R, +R, atbic atb+c atb+c 444 = b © a | =(a+b+q|beoa c a b cab Apply C,»C,~C,,C, »C,-C, o 0 1 =(atb+c)|b-c c-a a c-a a-b b = (a+b +6) ((b—c) (a—b)- (c—a)4) (a+b +0) (ab + be—ca—b?—o? + 2ca-a") = (a+b +0) (ab + be + ca—a?—b?— c?) = 3abe—a°— b*— abe Example #10 Simplify |a? b? c? be ca ab Solution: Given detereminantis equal to a po? a? pb? ? ecfe eo ]-88 lat eo Pe labe abe abc] Pf 1 1 1 Apply C,C,-C,, C,+C,-C, a? —b? b?-c? c? a—b? bec? o® o o 4 a+b bie? =(a—b)(b—c)]a’+ab+b? b?+bc+c? c® o 0 1 = (ab) (b— c) [ab? + abe + ac? +b’ + b°C + bc? — atb — atc — ab? - abo - b'— bc] (a -b) (b~c) [c(ab + be + ca) - alab + be + ca)] = (a—b) (b—c) (ca) (ab + be + ca) Self practice problems 0 b-a cua (5) Find the value of A=|a-b 0 cb a-c b-c 0 b?-ab b-c be-ac (6) Simplify }ab-a? a-b b?-ab be-ac c-a ab-a® pbc 2a 2a (7) Provethat] 22 P29 2D |= ays 2e 0 2e cabMatrices & Determinant @ 1a be (8) Showthat |1 b_ca| = (a—) (b-c) (ca) by using factor theorem 1 ¢ ab Answers: (5) 0 6) 0 Application of determinants : Following examples of short hand writing large expressions are: () Area ofa triangle whose vertices are (x, y,); r= 1, 2, Sis: x ys 4 Lhe yo 4 p=> pe % 1fD=0 then the three points are collinear. Ks Ys 4] xy (i) Equation of a straight line passing through (x,y) &(X, ¥2)i8 es Ya bk { (il) The fines: axtbyte, (1) 2% ax+by+o, (2) ax t by +c, = (3) lar by are concurrent if, lar b. | las bs cg Condition for the consistency of three simultaneous linear equations in 2 variables. (wy) ax? + 2hxy + by? + 29x + 2fy+6=0 represents a pair of straight lines if lah g =|h b ff bt ¢ abc + 2fgh — af? - bg*— ch* Singular & non singular matrix : A square matrix Ais said to be singular or non-singular according a8 [A|is zero or non-zero respectively. Cofactor matrix & adjoint matrix : Let A = [a], be a square matrix. The matrix obtained by replacing each element of A by corresponding cofactor is called as cofactor matrix of A, denoted as cofactor A. The transpose of cofactor matrix of A is called as adjoint of A, denoted as adj A. ie. HA then cofactor A ='[e,], when ¢, is the cofactor of a, vi &j AdjA= [dj], where d, = ¢) V8 Propertles of cofactor A and adj A; (a) A.adjA=|A|1, = (adj A) A where A= aj), (b) _[adjAl = |AI"~*; where nis order of A. In particular, for 3 x 3 matrix, |adj Al = |AI (c) If Ais a symmetric matrix, then adj A are also symmetric matrices. (a) If Ais singular, then adj Ais also singular. Example #11: For a 3x3 skew-symmetric matrix A, show that adj Ais a symmetric matrix, oO ab c? -be ca Solution A=|-a 0 ¢ cofA=|-bo b? —ab -b -c 0 ca -ab a? ce? -be ca -be b* —ab| which is symmetric. ca -ab a? adj A= (cof A)’ =Matrices & Determinant @ inverse of a matrix (recIprocal matrix) : 1 Let be a non-singular matrix. Then the matrix [7 adj Ais the multiplicative inverse of A (we call it inverse of A) and is denoted by A We have A (adj A) = |A| I, = (adj A) A : Al ) = mean Remarks: 4. Thenecessaryand suficient condton for exstence of verse of Ai hats non singula. 2. fis alvayenon-singulat 3. IAS dia (@4q, apy snes Apq) Where a, #0 V i, then AX = diag (ay,71, az9"1 4 (A-*y' = (A’)"* for any non-singular matrix A. Also adj (A’) = (adj Ay’. 5. (Atyt if Ais non-singular. 6. Let k be a non-zero scalar & Abe a non-singular matrix. Then (kA)? 1 7 IA“L= [ay for Al 0. Let A be a non-singular matrix. Then AB=AC => B=C & BA=CA=>B=C. a Ais non-singular and symmetric => A~' is symmetric. 10. (AB)"' = B* A“ if Aand B are non- singular. 41. In general AB = 0 does not imply A= 0 or B = 0. But if Ais non-singular and AB = 0, then B = 0 Sisinny nnonsingusrandAd woe Awe. Thertore,Abe acs einer bath ae lngosr orone femie® 413 3) cxamplen 12: tas" § 3], ten very tat Ag A= 1A Also Solution: We have |A| = 1 (16-9) 3 (4-3) +3 (3-4) =140 Now Cy, = 7, Cy == 1, Cyg =~ 1, Copy =~ 3, Czy = 1, Cry =0,Cy, 3, Cxy =0, Cag = 1 Therefore adj A= |~ 3] [7-3-3 -3+3+0 -3+0+9 Now A(adj A) = o| _|7-4-3 -3+4+0 -340+3 1] [7-3-4 -34340 -3.004 100 100 =f0 1 ey JO tO cya 001 °Matrices & Determinant @ 23 , where Lis 2 x 2 identity matrix and O is 2 x 2 zero matrix. Using the equation, find A 2 3\f2 3] [7 12 Solution WehaveA@=AA=1, oll, ol=l4 7 > 7 12 [8 12] [1 0 00 Hence ar—aa+i=[y a]-la e.tlo al=lo of =o Now Az 4A+ Therefore AA- 4A =-1 or AA(A“') 4 AA! or A(AR)— 4 1A+ (Post multiplying by A~t because [Al + 0) A or AI-4T=—At 40) [2 3] [2 -3] ai-a= lo gl-ly fel 2} | Example #14 : For two non-singular matrices A & B, show that adj (AB) = (adj B) (adj A) Solution: We have (AB) (adj (AB)) = [AB| I, = [AL Bl A“ (AB)(adj (AB)) = [A [8] A~* 1 = Badj (AB) = |B] adjA ic Tay as) = BB adj (AB) = |B| B~ adj A = adj (AB) = (adjB) (adj A) Self practice problems : (9) If Ais non-singular, show that adj (adj A) = |AI"~2A. (10) Prove that adj (A~") = (adj A)-? (11) For any square matrix A, show that Jadj (adj A)| = 1A [-" (12) IfAand B are non-singular matrices, show that (AB)-" = B~! A~" Elementary row transformation of matrix : The following operations on a matrix are called as elementary row transformations. (a) __Interchanging two rows, (bo) Multiplications of all the elements of row by a nonzero scalar. (c) Addition of constant multiple of a row to another row. Note : Similar to above we have elementary column transformations also, Remarks ; Two matrices A& B are said to be equivalentif one is obtained from other using elementary transformations, We write A= B. ing inverse using Elementry operations Using row transformations IfAis a matrix such that A exists, then to find A~’ using elementary row operations, Step 1: Write A= 1A and Step II : Apply a sequence of row operation on A = IA till we get, The matrix B will be inverse of A. Note : in order to apply a sequence of elementary row operations on the matrix equation X = AB, we will apply these row operatdions simultaneously on X and on the first matrix A of the product AB on RHS, daksnana BA.Matrices & Determinant (ii) ® Using column transformations : IfAis a matrix such that A~* exists, then to find A~! using elementary column operations, Step Step It Write A= Al and Apply a sequence of column operations on A= Al till we get, I = AB. The matrix B will be inverse of A, Note : In order to apply a sequence of elementary column operations on the matrix equation X = AB, we will apply these row operatdions simultaneously on X and on the second matrix B of the product AB on RHS. Example # 15 : Obtain the inverse of the matrix A= Solution or or or or Write A= 1A, ie., 2 S| using elementary operations 012) [100 12a|_lo10 Is 14} [0 0 1 3] foto 2|_|1 0 0 = A (applying R, <> R,) 1 oo1 ving 2 0 1 0 = 4 0 : A (applying R > Ry - 3R,) 10 0 Ol A (applying R, -» R, - 2R,) -31 0 ©! a (applying Ry > Ry + 5R,) o o 1 A (applying Ry -» Ry) 1 2 1] 2 © Ja (Applying R, -» R; +R) 4 2) {4 st 4) 0 22 2 ©) =|-4 3 -4]acAppying R, > R,-2R,) 1 § 4 2 2Matrices & Determinant @ System of IInear equations & matrices : Consider the system Aiyy Xq #4 Xy + + aaXy BK + gy Xp tH svseres By Ke BinaXt + BmaXe + + Bink + by ayy Be an x, by aor 22 aan Xe Let A= x=]? laps Amt Ae b b, Then the above system can be expressed in the matrix form as AX = B. The system is said to be consistent if it has atleast one solution System of IInear equations and matrix Inverse: If the above system consist of n equations inn unknowns, then we have AX = B where Ais a square matrix Results: (1)_If Ais non-singular, solution is given by X = A~'B. (2) IfAis singular, (adj A) B = 0 and all the columns of A are not proportional, then the system has infinitely many solutions. (3) IfAis singular and (adj) B + 0, then the system has no solution (we say it is inconsistent). Homogeneous system and matrix Inverse Ifthe above system is homogeneous, n equations in n unknowns, then in the matrix form itis AX = (in this case b, bb, = 0), where Ais a square matrix. Results: (1) _If Ais non-singular, the system has only the trivial solution (zero solution) X = 0 (2) If Ais singular, then the system has infinitely many solutions (including the trivial solution) and hence it has non-trivial solutions, xey+z=6 Example # 16 : Solve the system x-y+z=2 using matrix inverse, Dxey-z=t 104 x 6) Solution 1 y|ae= | 1 at Then the system is AX = B. IA] = 6. Hence A is non singular. 0 3 3] CofactorA=|2 -3 1 20Matrices & Determinant @ 02 2 aga=|3 -3 0 31-2 0.2 2] fo 13 43 agjasz|3 -3 0/=|12 -172 0 31-2) [2 v6 -13 0 13 113) [6 x) ft a2 -12 0 |l2] ic. |y|=|2 a2 6-13 [1 z| |s > KEL y=22=3, 12 (13) 2 3). Find the inverse of Ausing |A| and adj A. 1 (14) Find real values of 2. and 4 so that the following systems has (i) unique solution (ii) infinitely many solutions. (ii) No solution. xtytz x+2y+3z=4 x4 2y the (15) Find 2 so that the following homogeneous system have a non zero solution x + 2y + 3z= ix (14) (443, WE RGA=3,p=1 ()A=3, pet (15) R=6