Mathematics Foundations/6.2 Matrices and Matrix Operations

6.2 Matrices and Matrix Operations 6.2.1 Introduction to Matrices A matrix is a rectangular array of numbers, symbols, or expressions arranged in rows and columns. Matrices provide a compact way to represent and manipulate linear transformations, systems of linear equations, and other mathematical structures. 6.2.1.1 Matrix Notation An matrix has rows and columns: The entry in the -th row and -th column is denoted by or . 6.2.1.2 Special Types of Matrices Square Matrix: A matrix with the same number of rows and columns () Diagonal Matrix: A square matrix where all entries outside the main diagonal are zero Identity Matrix: A diagonal matrix with all diagonal entries equal to 1, denoted by Zero Matrix: A matrix with all entries equal to zero, denoted by Triangular Matrix: * Upper triangular: for all * Lower triangular: for all Symmetric Matrix: A square matrix equal to its transpose () Skew-Symmetric Matrix: A square matrix where Orthogonal Matrix: A square matrix whose inverse equals its transpose () 6.2.2 Matrix Operations 6.2.2.1 Matrix Addition and Scalar Multiplication For matrices and of the same dimensions : Addition: Scalar Multiplication: Properties: Commutativity of addition: Associativity of addition: Distributivity of scalar multiplication: and 6.2.2.2 Matrix Multiplication For an matrix and an matrix , their product is an matrix where: In other words, the entry in the -th row and -th column of is the dot product of the -th row of and the -th column of . Properties: Matrix multiplication is generally not commutative: in most cases Associativity: Distributivity: and Identity: for an matrix 6.2.2.3 Matrix Transpose The transpose of an matrix , denoted by , is the matrix obtained by interchanging the rows and columns of : Properties: 6.2.2.4 Trace The trace of a square matrix , denoted by , is the sum of its diagonal entries: Properties: 6.2.3 Matrix Inverses 6.2.3.1 Definition A square matrix is invertible (or nonsingular) if there exists a matrix such that: If such a matrix exists, it is called the inverse of . 6.2.3.2 Properties of Invertible Matrices For invertible matrices and : 6.2.3.3 Methods for Finding Inverses Method 1: Adjoint Method For a square matrix , the inverse is given by: where is the adjoint (or adjugate) of , which is the transpose of the matrix of cofactors. Method 2: Gaussian Elimination To find using Gaussian elimination: 1. Form the augmented matrix 2. Apply row operations to transform the left side to 3. The right side will become 6.2.3.4 Singular Matrices A square matrix that is not invertible is called singular. A matrix is singular if and only if its determinant is zero. 6.2.4 Elementary Matrices and Row Operations 6.2.4.1 Elementary Row Operations There are three types of elementary row operations: 1. Interchange two rows: 2. Multiply a row by a non-zero scalar: (where ) 3. Add a multiple of one row to another: (where ) 6.2.4.2 Elementary Matrices An elementary matrix is obtained by performing a single elementary row operation on the identity matrix. Multiplying a matrix on the left by an elementary matrix results in the same row operation being performed on : Properties: Every elementary matrix is invertible The inverse of an elementary matrix is also an elementary matrix 6.2.5 Rank and Nullity 6.2.5.1 Row and Column Spaces The row space of a matrix is the subspace of spanned by the rows of The column space of a matrix is the subspace of spanned by the columns of 6.2.5.2 Rank The rank of a matrix , denoted by , is the dimension of its column space (or equivalently, the dimension of its row space). Properties: for an matrix 6.2.5.3 Nullspace The nullspace (or kernel) of a matrix , denoted by or , is the set of all vectors such that : 6.2.5.4 Nullity The nullity of a matrix , denoted by , is the dimension of its nullspace. 6.2.5.5 Rank-Nullity Theorem For an matrix : 6.2.6 Determinants 6.2.6.1 Definition The determinant of a square matrix , denoted by or , is a scalar value that provides information about the matrix's invertibility and geometric properties. For a matrix: For a matrix: For larger matrices, the determinant can be calculated using cofactor expansion. 6.2.6.2 Properties of Determinants for an matrix If has a row or column of zeros, then If has two identical rows or columns, then If is triangular (upper or lower), then is the product of the diagonal entries if is invertible 6.2.6.3 Effect of Elementary Row Operations Interchanging two rows: Multiplying a row by a scalar : Adding a multiple of one row to another: remains unchanged 6.2.7 Matrix Factorizations 6.2.7.1 LU Decomposition An matrix can be factored as , where: is an lower triangular matrix with 1's on the diagonal is an upper triangular matrix This factorization is useful for solving systems of linear equations and computing determinants. 6.2.7.2 QR Decomposition An matrix can be factored as , where: is an orthogonal matrix () is an upper triangular matrix QR decomposition is commonly used in numerical methods for solving linear least squares problems. 6.2.7.3 Singular Value Decomposition (SVD) Any matrix can be factored as , where: is an orthogonal matrix is an diagonal matrix with non-negative real numbers (singular values) on the diagonal is an orthogonal matrix SVD has numerous applications in data analysis, image processing, and solving ill-conditioned systems. 6.2.8 Matrix Norms 6.2.8.1 Frobenius Norm The Frobenius norm of a matrix , denoted by , is defined as: 6.2.8.2 Spectral Norm The spectral norm (or 2-norm) of a matrix , denoted by , is defined as: where is the largest eigenvalue and is the largest singular value. 6.2.8.3 Induced Norms For , the induced p-norm of a matrix is defined as: Common induced norms: (maximum absolute column sum) (maximum absolute row sum) 6.2.9 Exercises 1. Let and . Calculate and . 2. Find the inverse of the matrix using the adjoint method. 3. Determine the rank and nullity of the matrix . 4. Find the determinant of using cofactor expansion. 5. Perform an LU decomposition of the matrix . 6. Prove that if is a skew-symmetric matrix (i.e., ), then all its diagonal entries are zero.