QUANTITATIVE BOOTCAMP

for the Masters of Science in Business Analytics

(IEOR E4599)
LECTURE #1
OUTLINE

 Class #1: Linear algebra

 Class #2: Calculus
 Class #3: Statistics
OVERVIEW

 Purpose of this lecture series

 Additional resources
 Online: www.sosmath.com
 Mathematics & Statistics for Financial Risk Management
 A Primer for the Mathematics of Financial Engineering
 Mini-Quizzes
 CourseWorks (aka Canvas)
 Very easy or very difficult?
 [email protected]
LINEAR ALGEBRA

 Matrix operations and properties

 Matrix inversion
 Vectors
 Inner product
 Norm
 Linear independence
 Rank of a matrix
 Eigenvectors and eigenvalues
MATRIX OPERATIONS AND PROPERTIES

Matrix Notation

𝑎1,1 𝑎1,2 4 9 3
𝐂=
𝐛= 7 5 3
𝐀 = 𝑎2,1 𝑎2,2
𝑎3,1 𝑎3,2 −2 2 x 2 square matrix
3 x 1 vector
3 x 2 matrix

−5 0 0 0
10 −2 6
0 0 0 0 𝐄 = −2 5
𝐃= 3
0 0 3 0
6 3 0
0 0 0 7
3 x 3 symmetrical matrix
4 x 4 diagonal matrix
MATRIX OPERATIONS AND PROPERTIES

Addition, subtraction, multiplication

𝑎 𝑏 𝑒 𝑓 (𝑎 + 𝑒) (𝑏 + 𝑓) 𝑎 𝑏 𝑘𝑎 𝑘𝑏
+ = 𝑘
𝑐 𝑑
=
𝑘𝑐 𝑘𝑑
(scalar multiplication)
𝑐 𝑑 𝑔 ℎ (𝑐 + 𝑔) (𝑑 + ℎ)

𝑎 𝑏 𝑒 𝑓 (𝑎 − 𝑒) (𝑏 − 𝑓) 𝑔 ℎ
− = 𝑎 𝑏 𝑐 (𝑎𝑔 + 𝑏𝑖 + 𝑐𝑘) (𝑎ℎ + 𝑏𝑗 + 𝑐𝑙)
𝑐 𝑑 𝑔 ℎ (𝑐 − 𝑔) (𝑑 − ℎ) 𝑖 𝑗 =
𝑑 𝑒 𝑓 (𝑑𝑔 + 𝑒𝑖 + 𝑓𝑘) (𝑑ℎ + 𝑒𝑗 + 𝑓𝑙)
𝑘 𝑙

For addition: For multiplication: 𝑛

• Amxn + Bmxn= Cmxn • AmxnBnxp= Cmxp 𝑐𝑟,𝑐 = ෍ 𝑎𝑟,𝑖 𝑏𝑖,𝑐

𝑖=1
• A + B = B +A • AB ≠ BA
MATRIX OPERATIONS AND PROPERTIES

Transpose Identity matrix and inversion

𝐀′ or 𝐀T
1 0 0
𝐈= 0 1 0
If 0 0 1
1 2 1 2
𝐀=
3 4 𝐁= 3 4 𝐀𝐈 = 𝐀 = 𝐈𝐀
5 6
Then
𝐀𝐀−1 = 𝐀−1 𝐀 = 𝐈
1 3 1 3 5
𝐀′ = 𝐁′ =
2 4 2 4 6

More formally
a′𝑖,𝑗 = 𝑎𝑗,𝑖
MATRIX INVERSION

 Need to be comfortable with 2x2 and 3x3

 2x2 is easy
 3x3 is a bit tedious
 Anything larger and you probably want a computer
MATRIX INVERSION

2x2
In general

𝑎 𝑏 1 𝑑 −𝑏 1 𝑑 −𝑏
𝐀= 𝐀−𝟏 = =
𝑐 𝑑 det(A) −𝑐 𝑎 𝑎𝑑 − 𝑐𝑏 −𝑐 𝑎
Example

4 2 1 3 −2 1 3 −2 3/2 −1
𝐀= 𝐀−𝟏 = = =
5 3 4 × 3 − 5 × 2 −5 4 2 −5 4 −5/2 2
MATRIX INVERSION

Two methods
 Gauss-Jordan elimination  Minors, cofactors, adjoints and
determinants
4 2 1 0
ቚ
5 3 0 1
(multiply rows by
Example online: MatrixInverse.3x3.pdf
⋮ constant, then
add/subtract/swap)

1 0 3/2 −1
ฬ
0 1 −5/2 2
MINI-QUIZ #1
VECTORS

Inner product or dot product

𝐚 ∙ 𝐛 = 𝑎1 𝑏1 + 𝑎2 𝑏2 + ⋯ + 𝑎𝑛 𝑏𝑛

𝜃
Norm, magnitude, or length

𝐯 = 𝐯∙𝐯

𝐚∙𝐛= 𝐚 𝐛 cos(𝜃)
Orthogonal vectors
𝐚∙𝐛=0
VECTORS

Linear independence
𝑣1 𝑤1 𝑥1
𝑣2 𝑤2 𝑥2
c1 ⋮ + c2 ⋮ + c3 ⋮ = 𝟎 𝑖𝑓𝑓 c1 = c2 = c3 = 0
𝑣𝑛 𝑤𝑛 𝑥𝑛

• Orthogonal → independent
? orthogonal
• Independent →
• Changing the length of vectors doesn’t
change independence
• Rotation does not change independence
MINI-QUIZ #2
VECTORS

Vector space
v1 dot v2 has to be orthogonal, i.e. v1 dot v2 = 0
 A collection of vectors
 Examples: R2, R3

Basis
 All the vectors in a vector space can be
defined in terms of a linear combination of
the vectors in the basis for that space.
 The vectors in the basis are independent.
 Orthonormal basis:
 all vectors in basis are orthogonal and of unit
length (norm = 1)
 𝐕 ′ 𝐕 = 𝐈 = 𝐕 −1 𝐕
MATRIX RANK

Given a matrix, how many dimensions

does it span?

1/ 2 −1/ 2
1 0 1 3
0 1 1/ 2 1/ 2 1 3
Rank = 2 Rank = 2 Rank = 1
MATRIX RANK

For any matrix:

 Rank(A) ≤ min(m, n)
 Rank(A) = Rank(A')

𝑎1,1 ⋯ 𝑎1,𝑛  If Rank(A) = min(m, n), then A is of full rank

𝐀= ⋮ ⋱ ⋮
𝑎𝑚,1 ⋯ 𝑎𝑚,𝑛 For a square matrix, the following are
equivalent:
 A has full rank
 A is invertible
 The column vectors of A are linearly independent
 The row vectors of A are linearly independent
MINI-QUIZ #3
EIGENVALUES & EIGENVECTORS

Definition Example
Given:
𝐀𝐯 = 𝜆𝐯 3 0 1
𝐀= 𝐯 = 𝜆=3
𝐀 − 𝜆𝐈 𝐯 = 𝟎 8 −1 2
We have:
• A is a square matrix 3 0 1 3 1
𝐀𝐯 = = =3 = 𝜆𝐯
• λ is an eigenvalue 8 −1 2 6 2
or,
• v is the corresponding eigenvector 3 0 1 0 1
𝐀 − 𝜆𝐈 𝐯 = −3
8 −1 0 1 2
3 0 3 0 1
= −
8 −1 0 3 2
0 0 1
=
8 −4 2
0
=
0
=𝟎
MINI-QUIZ #4