LINEAR ALGEBRA

Alessandra Buratto

Department of Mathematics, University of Padua

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
A brush up on matrices

   
1 0 3 1 0 3 3
A = 2 3 1  B = 2 3 1 1
0 −2 2 0 −2 2 2

See video “Introduction to Matrices” in Moodle

Simon Blume pages 153–173

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
 DETERMINANTS: An overview

Let be Mn,n the set of nxn SQUARE matrices (n rows and n

columns), let ’s consider in it the real value function:

Determinant := function Mn,n → IR

First and Second order definition

n=1 det(a) = a
 
a11 a12
n=2 det = a11 a22 − a21 a12
a21 a22

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Some examples

√ √
  √ 
1 4 2 0
det = 3 − 8 = −5, det = 2 − 0 = 2,
2 3 0 1
   
1 1 1 4
det = 2 − 2 = 0, det = 8 − 8 = 0.
2 2 2 8

If det(a) = 0 the matrix A is called singular

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Minors: Definition
Definition (Minor of order k)
Let be B a mxn matrix. Let B̃ any k x k square submatrix
obtained considering k rows and k columns of B. Then the
following scalar is called the minor of B of order k
det B̃
The order of a minor is the number of its rows (columns)
Definition (Minor of element aij )
Let be A a nxn square matrix of order n ≥ 2 and let aij one of its
elements (in the i−th row and j−th column). Let Aij the
(n − 1)x(n − 1) submatrix obtained by deleting row i and column j
from matrix A. Then the following scalar is called the (i, j)th minor
of A
Mij = det Aij
Alessandra Buratto Department of Mathematics, University of Padua
LINEAR ALGEBRA
Example of minors of elements

Given the matrix  

1 0 3
A = 2 3 1 
0 −2 2
   
0 3 1 0
M21 = Det = 6, M33 = Det =3
−2 2 2 3

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Determinant: dimension 2

Let’s observe that if n = 2

 
a11 a12
a21 a22

the determinant can be rewritten in the following way

det(A) = a11 a22 − a21 a12 = a11 M11 − a21 M21 = a11 M11 − a12 M12 =
= −a12 M12 + a22 M22 = −a21 M21 + a22 M22 .

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Determinant: general rule
Definition (Determinant)
let be A a nxn n ≥ 2, then, for any fixed j ∈ {1..n} we define
(column development)
n
X
det A = (−1)i+j aij Mij . (1)
i=1

Observation The Definition can be developed starting from any

column of A .
n
X
det A = (−1)i+j aij Mij . (2)
j=1

comment on Cofactor
Alessandra Buratto Department of Mathematics, University of Padua
LINEAR ALGEBRA
 Example

 
1 0 3      
3 1 0 3 0 3
det 2 3 1 = 1 det −2 det +0 det =
−2 2 −2 2 3 1
0 −2 2

= 1(6 + 2) − 2(0 + 6) + 0(0 − 9) = 8 − 12 + 0 = −4.

WARNING: Sarrus formula is applicable only for 3x3 matrices

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Elementary cases

   
0 0 0 1 0 0
det 0 = det 0 0 0 = 0 det I3 = det 0 1 0 = 1
0 0 0 0 0 1

det A = det AT (AT transpose of A)

If A has one row/ column of zeros ⇒ det A = 0
If A has rows / columns linearly DEPENDENT ⇒ det A = 0

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Properties -1

If A is DIAGONAL
 
a11 0 0
n
 0 a22 0  Y
A=  ⇒ det A = aii
. . . . . . . . . 
i=1
0 0 ann
Qn
If A is TRIANGULAR ⇒ det A = i=1 aii
If A e B are both nxn matrices then AB is nxn and

det AB = det A · det B (generally det AB 6= det BA)

det AB 6= 0 ⇔ det A 6= 0 and det B 6= 0

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Inverse matrix
Let be A a nxn square matrix of order n ≥ 2 with det A 6= 0
A is invertible if there exists a nxn square matrix (A−1 ) such that

A · A−1 = A−1 · A = 1 = In
A is invertible ⇔ det A 6= 0 (rows/columns are
linearly independent)
1 = det 1 = det A · A−1 = det A · det A−1

and its inverse matrix is is

A−1 = (bij )
where
Mji
bij = (−1)i+j
det A
Alessandra Buratto Department of Mathematics, University of Padua
LINEAR ALGEBRA
Principal minors

Definition (Principal Minor of order k)

Let be A, an n × n, squared matrix. A given k × k submatrix of A
formed by deleting n − k columns, and the same n − k rows from
A is called principal submatrix of A. The determinant of a k × k
principal submatrix is called principal minor of A.
Observation
k ≤ min{m, n}

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
 Principal minors: Examples
 
1 2 4
A = 3 1 2 
4 3 6
order k = 1 : 1, 1, 6
order k = 2 :
 
1 2
Rows{1, 2} columns {1, 2} det = −5
3 1
 
1 4
Rows{1, 3} columns {1, 3} det = −10
4 6
 
1 2
Rows{2, 3} columns {2, 3} det =0
3 6
order k = 3 Rows {1, 2, 3} columns {1, 2, 3} : det(A) = 0
Alessandra Buratto Department of Mathematics, University of Padua
LINEAR ALGEBRA
Rank

Definition (Rank)
Let be A, a mxn matrix. We call rank of A the maximum order of
its non null minors.

rank(A) is the maximum number

of rows (columns) linearly independent

If rank(A) = m − 1 then the rows of A are linearly dependent.

If rank(A) = n − 1 then the columns of A are linearly dependent.

rank(A) = rank(AT )
If A ∈ Mn,n is n × n then A is invertible if and only if its rank is n.

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Rank: Example
 
1 2 −1 4
A = 3 1 2 2 
4 3 1 6
 
1 2
Order 2 minor (rows (1,2) columns (1,2)) det = −5 6= 0
3 1
rank(A) ≥ 2.

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Rank: Example
 
1 2 −1 4
A = 3 1 2 2 
4 3 1 6
 
1 2
Order 2 minor (rows (1,2) columns (1,2)) det = −5 6= 0
3 1
rank(A) ≥ 2. Go on analysing order 3 minors as long as they are
null    
1 2 −1 1 2 4
det 3 1 2  = 0 det 3 1 2 = 0
4 3 1 4 3 6
   
1 −1 4 2 −1 4
det 3 2 2 = 0 det 1 2 2 = 0
4 1 6 3 1 6
All the order 3 minors are null, so that rank(A) = 2
Alessandra Buratto Department of Mathematics, University of Padua
LINEAR ALGEBRA
Leading Principal Minor: Example
Definition (Principal Minor)
Let be A an nxn matrix. We call Principal Minor of order k the
determinant of the order k submatrix obtained by A deleting n − k
rows and n − k columns.
 
1 0 3
A = 2 3 1
0 −2 2

p.m. order 1 = diagonal elements A = 1, 3, 2

p.m. order 2
     
1 0 1 3 3 1
=3 det =2 det =8
2 3 0 2 −2 2

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Principal Minor: Example

p.m. order 3= matrix determinant

 
1 0 3
det 2 3 1 = −4
0 −2 2

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA
Leading Principal Minor
Let be A an nxn matrix. We call Leading Principal Minor of order
k the determinant of the order k submatrix obtained by A deleting
the last n − k rows and the last n − k columns.
 
1 0 3
A = 2 3 1
0 −2 2
l.p.m. order 1: a11 = 1
 
1 0
l.p.m. order 2: det =3
2 3
 
1 0 3
l.p.m. order 3: matrix determinant det 2 3 1 = −4
0 −2 2

Alessandra Buratto Department of Mathematics, University of Padua

LINEAR ALGEBRA