Section 3.

3: Cramer’s Rule, Volume, and

Linear Transformations
Cramer’s Rule (1 of 4)
• Theorem 7: Let A be an invertible n  n matrix. For any b in n
,
the unique solution x of Ax = b has entries given by

det Ai ( b )
xi = , i = 1, 2 , , n (1)
det A

where Ai (b) is formed by replacing column i of A by b.

• Proof Denote the columns of A by a1 , . . . , a n and the columns of the

n  n identity matrix I by e1 , ,e n .
Cramer’s Rule (2 of 4)
• If Ax = b, the definition of matrix multiplication shows that

AI i ( x ) = A e1 x e n  =  Ae1 Ax Ae n 

= a1 b a n  = Ai ( b )

• By the multiplicative property of determinants,

(det A ) (det I i (x)) = det Ai (b)

• The second determinant on the left is simply xi . Hence

(det A) xi = det Ai (b). This proves (1) because A is invertible and det A ≠ 0.
Cramer’s Rule (3 of 4)
• Example 1 Use Cramer’s rule to solve the system
3 x1 − 2 x2 = 6
−5 x1 + 4 x2 = 8
• Solution View the system as Ax = b. Using the notation
3 − 2
introduced above, A=  ,
 −5 4 
6 − 2 3 6
A1 ( b ) =   , A2 ( b ) = 
8 4   −5 8
• Since det A = 2, the system has a unique solution.
Cramer’s Rule (4 of 4)
• By Cramer’s rule,

det A1 ( b ) 24 + 16
x1 = = = 20
det A 2
det A2 ( b ) 24 + 30
x2 = = = 27
det A 2
A Formula For A to the Negative First Power (1 of 4)
• The adjugate matrix is the transpose of the matrix of cofactors.

• Theorem 8: Let A be an invertible n  n matrix. Then

1
A =
−1
adj A
det A
A Formula For A to the Negative First Power (2 of 4)
2 1 3 
• Example 3 Find the inverse of the matrix A = 1 − 1 1  .
 
• Solution The nine cofactors are
1 4 − 2 

−1 1 1 1 1 −1
C11 = + = −2, C12 = − = 3, C13 = + =5
4 −2 1 −2 1 4
1 3 2 3 2 1
C21 = − = 14, C22 = + = −7, C23 = − = −7
4 −2 1 −2 1 4
1 3 2 3 2 1
C31 = + = 4, C32 = − = 1, C33 = + = −3
−1 1 1 1 1 −1
Thus
 −2 14 4 
adj A = 3 − 7 1 
 
5 − 7 − 3
A Formula For A to the Negative First Power (3 of 4)
• We could compute det A directly, but the following computation
provides a check on the calculations above and produces det A:

 −2 14 4  2 1 3 
( adj A) A = 3 −7 1  1

−1 1

5 −7 − 3 1 4 − 2 

14 0 0
= 0 14 0  = 14 I
 
0 0 14 
A Formula For A to the Negative First Power (4 of 4)
• Since (adj A) A = 14 I, Theorem 8 shows that det A = 14 and

 −2 14 4   −1/ 7 1 2/7 
1 
A =
−1
3 −7 1  = 3 /14 − 1/ 2 1/14 
14    
5 −7 − 3 5 /14 − 1/ 2 − 3 /14 
Determinants as Area or Volume (1 of 9)
• Theorem 9: If A is a 2  2 matrix, the area of the parallelogram
determined by the columns of A is det A .

• If A is a 3  3 matrix, the volume of the parallelepiped determined

by the columns of A is det A .

• Proof The theorem is obviously true for any 2  2 diagonal matrix:

a 0
det   = ad = {area of rectangle}
0 d
Determinants as Area or Volume (2 of 9)

• It will suffice to show that any 2  2 matrix A = a1 a 2 

can be transformed into a diagonal matrix in a way that changes
neither the area of the associated parallelogram nor det A .
Determinants as Area or Volume (3 of 9)
• It suffices to prove the following simple geometric observation that
applies to vectors in 2
or 3
:
• Let a1 and a 2 be nonzero vectors. Then for any scalar c, the area of
the parallelogram determined by a1 and a 2 equals the area of the
parallelogram determined by a1 and a 2 + Ca1 .

• To prove this statement, we may assume that a 2 is not a multiple of

a1 , for otherwise the two parallelograms would be degenerate and
have zero area.
• If L is the line through 0 and a1 , then a 2 + L is the line through a2
parallel to L, and a 2 + ca1 is on this line.
Determinants as Area or Volume (4 of 9)

• The points a 2 and a 2 + ca1 have the same perpendicular

distance to L. Hence the two parallelograms have the
same area, since they share the base from 0 to a1 .
Determinants as Area or Volume (5 of 9)
• The proof for 3
is similar. The theorem is obviously true
for a 3  3 diagonal matrix.
Determinants as Area or Volume (6 of 9)
• And any 3  3 matrix A can be transformed into a diagonal
matrix using column operations that do not change det A .

• A parallelepiped is shown below as a shaded box with two

sloping sides.
Determinants as Area or Volume (7 of 9)
• Its volume is the area of the base in the plane Span {a1 , a3 }
times the altitude of a 2 above Span {a1 , a3 }. Any vector a 2 + ca1
lies in the plane Span {a1 , a3 }.

• Hence the volume of the parallelepiped is unchanged when

a1
a2 a3  is changed to a 1
a 2 + ca1 a3 .

• Thus a column replacement operation does not affect the volume of the
parallelepiped. Since the column interchanges have no effect on the volume,
the proof is complete.
Determinants as Area or Volume (8 of 9)
• Example 4 Calculate the area of the parallelogram
determined by the points ( −2, −2 ) , ( 0,3) , ( 4, −1) , and ( 6, 4 ) .
Determinants as Area or Volume (9 of 9)
• Solution First translate the parallelogram to one having the
origin as a vertex. For example, subtract the vertex
( −2, −2 ) from each of the four vertices.

• The new parallelogram has the same area, and its vertices are
(0, 0), (2, 5), (6, 1), and (8, 6).
• This parallelogram is determined by the columns of

2 6
A= 
 5 1 
• Since det A = −28 , the area of the parallelogram is 28.
Linear Transformations (1 of 5)
• Theorem 10: Let T: 2
→ 2
be the linear transformation
determined by a 2  2 matrix A. If S is a parallelogram in 2
, then

{area of T ( S )} = det A  {area of S} (5)

• If T is determined by a 3  3 matrix A, and if S is a parallelepiped in
3
, then
{volume of T ( S )} = det A  {volume of S} (6)
• Proof Consider the 2  2 case, with A = a1 a 2 .A
parallelogram at the origin in 2
determined by vectors b1
and b 2 has the form
S = {s1b1 + s2 b 2 : 0  s1  1, 0  s2  1}
Linear Transformations (2 of 5)
• The image of S under T consists of points of the form

T ( s1b1 + s2 b 2 ) = s1T ( b1 ) + s2T ( b 2 )

= s1b1 + s2 b 2

• where 0  s1  1, 0  s 2  1. It follows that T(S) is the parallelogram

determined by the columns of the matrix  Ab1
Ab 2 .

This matrix can be written as AB, where B = b1 b 2 .

• By Theorem 9 and the product theorem for determinants,
{area of T ( S )} = det AB = det A det B

= det A {area of S} (7)

Linear Transformations (3 of 5)
• An arbitrary parallelogram has the form p + S, where p is a vector and S is a
parallelogram at the origin.
• It is easy to see that T transforms p + S into T (p ) + T ( S ).
Since translation does not affect the area of a set,

{area of T ( p + S )} = {area of T ( p ) + T ( S )}
= {area of T ( S )} Translation
= det A {area of S} By equation ( 7 )
= det A {area of p + S} Translation

• This shows that (5) holds for all parallelograms in 2

.
The proof of (6) for the 3  3 case is analogous.
Linear Transformations (4 of 5)
• Example 5 Let a and b be positive numbers. Find the area of the region E
bounded by the ellipse whose equation is

2 2
x x
1 2
2
+ 2
=1
a b

• Solution We claim that E is the image of the unit disk D under the linear
transformation T determined by the matrix

a 0 u1   x1 
A=  , becauce if u =   , x =   , and x = Au, then
0 b  u2   x2 

x1 x
u1 = and u2 = 2
a b
Linear Transformations (5 of 5)
1 2
• It follows that u is in the unit disk, with u + u  1,
2 2

(x / a ) + ( x2 / b )  1.
2 2
if any only if x is in E, with 1

• By generalization of Theorem 10,

{area of ellipse} = {area of T ( D )}

= det A {area of D}

= ab  (1) =  ab
2
 Linear Algebra and Its Applications

Chapter 4
Section 4.1: Vector Spaces and Subspaces
Section: 4.1: Vector Spaces and Subspaces
Vector Spaces (1 of 7)
• Definition: A vector space is a nonempty set V of objects,
called vectors,
• on which are defined two operations, called addition and
multiplication by scalars (real numbers), subject to the ten
axioms listed on the next slide.
• The axioms must hold for all vectors u, v, and w in V and for
all scalars c and d.
Vector Spaces (2 of 7)
1. The sum of u and v, u + v is in V.

2. u+v =v+u

3. (u + v) + w = u + (v + w )
4. There is a zero vector in V such that u + 0 = u.

5. For each u in V, there is a vector −u in V such that u + ( − u ) = 0.

6. The scalar multiple of u by c, cu, is in V.

7. c ( u + v ) = cu + cv 9. c ( du ) = ( cd ) u

8. ( c + d ) u = cu + cv 10. 1u = u
Vector Spaces (3 of 7)
• Using these axioms, we can show that the zero vector is unique.
• The vector −u, called the negative of u, is also unique for
each u in V.
• The set n
is a vector space under the usual addition and
scalar multiplication.
Vector Spaces (4 of 7)
• For each u in V and scalar c,

0u = 0
c0 = 0
−u = ( −1)u
Vector Spaces (5 of 7)
• Example 1: Let V be the set of all arrows (directed line segments) in three-
dimensional space, with two arrows regarded as equal if they have the same
length and point in the same direction. Define addition by the parallelogram
rule, and for each v in V, define cv
to be the arrow whose length is c times the length of v, pointing in
the same direction as v if c  0 and otherwise pointing in the
opposite direction.
Vector Spaces (6 of 7)

• The definition of V is geometric, using concepts of length and

direction.
• No xyz-coordinate system is involved.
• An arrow of zero length is a single point an represents the zero
vector.

• The negative of v is ( −1) v.

Vector Spaces (7 of 7)
• Axioms 1, 4, 5, 6, and 10 are evident.
 Vector Spaces
Practice: Let V be the first quadrant in the xy-plane; that is, let
𝑥
V = 𝑦 : 𝑥 ≥ 0, 𝑦 ≥ 0

a) If u and v are in V, is u + v in V? why?

b) Find a specific vector u in V and a specific scalar c such that cu is not in V.
(This is enough to show that V is not a vector space.)
Solutions to Practice:
a) If u and v are in V, then their entries are nonnegative. Since a sum of
nonnegative numbers is nonnegative, the vector u + v has nonnegative
𝑥1 𝑦1 𝑥 +𝑦
entries. Thus u + v is in V. such that u = 𝑥 , u = 𝑦 , then u +v = 𝑥1 + 𝑦1
2 2 2 2
2
b) Example: If u = , and c = -1, then u is in V but cu is not in V.
2
Vector Spaces
Practice: Let W be the union of the first and third quadrants in the xy-plane.
𝑥
That is, let W = 𝑦 : 𝑥𝑦 ≥ 0
a) If u is in W and c is any scalar, is cu in W? Why?
b) Find specific vectors u and v in W such that u + v is not in W. (This is enough to show that W is
not a vector space.)
Solutions to Practice:
𝑥 𝑥 𝑐𝑥
a) If u = 𝑦 is in W, then the vector cu = c 𝑦 = 𝑐𝑦 is in W because (cx)(cy)= c 2 xy ≥ 0 since xy ≥
0.
−1 2
b) Example: If u = , and 𝐯 = ,.
−7 3
−1 2 1
u+v= + = ,
−7 3 −4
then u and v are in W but u + v is not in W
xy = (1 x (-4)) = -4 < 0, it is not in 𝑾.
The Polynomials of Degree at Most n
• For n  0, the set n of polynomials of degree at most n, is a
vector space.
• It consists of all polynomials of the form

p ( t ) = a0 + a1 t + a2 t 2 +  + an t n

where the coefficients a0 ,..., an and the variable t are real numbers.

• The degree of p is the highest power of t whose coefficient is not zero. If all
the coefficients are zero, p is called the zero polynomial.
Subspaces (1 of 3)
• Definition: A subspace of a vector space V is a subset H of V that
has three properties:
a. The zero vector of V is in H.
b. H is closed under vector addition. That is, for
each u and v in H, the sum u + v is in H.
c. H is closed under multiplication by scalars. That is, for
each u in H and each scalar c, the vector cu is in H.
Subspaces (2 of 3)
• Properties (a), (b), and (c) guarantee that a subspace H of V is itself a vector space,
under the vector space operations already defined in V.
• Every subspace is a vector space.
• Conversely, every vector space is a subspace (of itself and possibly of other larger
spaces).
Subspaces (3 of 3)
• The set consisting of only the zero vector in a vector space V is
a subspace of V, called the zero subspace and written as {0}.
• Recall that the term linear combination refers to any sum of scalar
multiples of vectors, and

Span  v1 ,..., v p 

• denotes the set of all vectors that can be written as linear

combinations of v1 ,...v p .
A Subspace Spanned by a Set (1 of 2)
• Example 2: Given v1 and v 2 in a vector space V, let H = Span v1 , v 2 .
Show that H is a subspace of V.

• Solution:
a. The zero vector is in H, since 0 = 0v1 + 0v 2 .
b. Take two arbitrary vectors in H, say

w = t1 v1 + t2 v 2 and u = s1 v1 + s2 v 2 , then

u + w = ( s1 + t1 ) v1 + ( s2 + t2 ) v 2 is in H.

c. If c is any scalar, cu = ( cs1 ) v1 + ( cs2 ) v 2 is in H.

Thus H is a subspace of V.
A Subspace Spanned by a Set (2 of 2)
• Theorem 1: If v1 ,..., v p are in a vector space V, then
Span v1 ,..., v p  is a subspace of V.

• We call Span v ,..., v  the subspace spanned (or generated) by

1 p

v ,..., v .
1 p

• Give any subspace H of V, a spanning (or generating) set for H is a

set v ,..., v  in H such that
1 p

H = Span v1 ,..., v p .

A Subspace Spanned
Practice: Show that the set H of all points in 𝑹𝟐 of the form (3s, 2 + 5s)
is not a vector space, by showing that it is not closed under scalar
multiplication. (Find a specific vector u in H and a scalar c such that
cu is not in H.)
Solutions to Practice:
3
Take any u in H - say, u = , and take any c ≠1, say, c = 2. Then
7
6
cu = . If this is in H , then there is some s such that
14
3𝑠 6
=
2 + 5𝑠 14
That is, s = 2 and s = 12/ 5, which is impossible. So 2u is not in H
and H is not a vector space
A Subspace Spanned
𝑠
Practice: Let H be the set of all vectors of the form 3𝑠 . Find vector v in
2𝑠
𝑹𝟑 such that H = Span {v}.

Solutions to Practice:
0 1 𝑠 𝑇 𝑠+𝑇
Let s=0, the v= 0 . cv = cs 3 . S = 3𝑠 , T= 3𝑇 = S+T = 3(𝑠 + 𝑇)
0 2 2𝑠 2𝑇 2(𝑠 + 𝑇)
𝟏
The set H = Span{v}, where 𝐯 = 𝟑 . Thus H is a subspace of 𝑹𝟑 by
𝟐
Theorem 1