0% found this document useful (0 votes)
25 views16 pages

ECON2272: Matrix Operations and Theory

The document outlines tutorial exercises for an Intermediate Mathematics for Economists course, covering topics such as matrix operations, linear independence, systems of equations, and applications in economics. Each tutorial consists of multiple questions that require solving mathematical problems related to matrices, determinants, and regression models. The exercises are designed to enhance understanding of mathematical concepts applicable to economic analysis.

Uploaded by

modexlurr
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
25 views16 pages

ECON2272: Matrix Operations and Theory

The document outlines tutorial exercises for an Intermediate Mathematics for Economists course, covering topics such as matrix operations, linear independence, systems of equations, and applications in economics. Each tutorial consists of multiple questions that require solving mathematical problems related to matrices, determinants, and regression models. The exercises are designed to enhance understanding of mathematical concepts applicable to economic analysis.

Uploaded by

modexlurr
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Business School

Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)


Tutorial 1. Basic Matrix Operations

Solve all questions. A selection of questions will be marked.


Question 1
Consider the following matrices

 3 5  7 3
A  , B 
4 1  0 2
a) Work out AB and BA. Comment
b) Work out tr(AB) and tr(BA). Comment

Question 2
 5 0 1 
3 1 0  
Let A    and B   2 3 7
 5 2 8   6 1 4 
 
a) Work out AB. Does BA. Exist?
b) Work out (AB)’ and B’A’. Comment
c) Verify that AA’ is symmetric

Question 3
6 9 1 3  4 2
Let A   , B    , and C 
4 6 0 1 6 3
Find A²,B² and C²

Question 4
xx
Let x be n×1 matrix whose j-th element is x j . Show that A  and B  I n  A are
xx
symmetric idempotent matrices.
Hint: Note that x x is a scalar (real number).

1
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 2. Linear Independence of Vectors, Rank and Inverse of a Matrix

Solve all questions. A selection of questions will be marked.


Question 1
1   3
a) Prove that the vectors   ,   are linearly independent.
2
  1 

1   3 9
b) The vectors u =   , v =   , w =   are linearly dependent. Which linear combination
2
   1 8
 
of u and v gives w? Illustrate your answer with a graph.

Question 2
a) Evaluate |A|, where
 2 5 4 7
 
3 8 6 0
A
 1 3 2 4
 
 4 1 8 3 
Hint: Adding a multiple of one column to another column does not change the determinant.
Create a column with as many zeros as possible before evaluating he determinant.
b) Does matrix A have an inverse?

Question 3
Use elementary row operations to reduce A to its echelon form and determine its rank.
 7 1 0 
 
A 1 1 4
 13 3 4 
 
Check your result using determinants.

Question 4
1 2 1 

Let A  1 3 2

 
1 0 1 
 
Prove that A is nonsingular and find its inverse:
a) Using cofactors.
b) Using elementary row operations.
2
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 3. Linear Simultaneous Equations

Solve all questions. A selection of questions will be marked.

Question 1
Consider the homogenous system of linear equations:
x1  3x3  0
4 x1  2 x2  2 x3  0
 x1  x2  7 x3  0
a) For what values of  will the system have only the trivial solution?
b) For what values of  will the system have an infinite number of solutions? Give the form
of the general solution in the case.

Question 2
Consider the non-homogenous system of linear equations:
18 x1  5 x2  23
6 x1  2 x2  8
9 x1  3x2  12
Show that this system is consider (i.e. a solution exists) and the solutions is unique. Find the
solution.

Question 3
Consider the system of equations
x1  2 x2  2 x3  5
 x1  2 x2  x3  1
4 x1  8 x2  5 x3  8
3x1  6 x2  x3  5
a) Show that this system has an infinite number of solutions
b) A theorem states that the general solution to the non-homogenous system Ax = b is equal
to a particular (special) solution to Ax = b plus the general solution to the homogeneous system
Ax = 0. Use this theorem to find the general solution to the system of equations.

3
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 4. Applications

Solve all questions. A selection of questions will be marked.

Question 1
General equilibrium analysis considers the interaction between markets. For example, an
increase in the price of silver may affect the market for lead because silver and lead are often
found in the same geological formations. Consider the following demand and supply functions
for the two metals:

Demand for Sliver: S  a0  a1 PS


d

Supply of sliver: S S  b0  b1 PS  b2 Pl

Demand for Lead: Ld   0  1Pl  ul

Supply of lead: LS  0  1PS  2 Pl

The term ul indicates an exogenous shock to the demand for lead.

a) What are the signs of b2 and 1 if silver and lead are joint products?

b) In a general equilibrium demand equals supply in each market.

a0  a1PS  b0  b1PS  b2 Pl

 0  1Pl  ul  0  1PS  2 Pl
This can be written as:

 a1  b1  PS  b2 Pl  b0  a0
 1 PS  1   2  Pl    0   0   ul
Use Cramer’s rule to solve this system of equations for the price of silver.
c) Suppose the shift to alternative energy sources causes an exogenous increase in the demand

for batteries that increases the demand for lead ( ul  0 ).What is the effect on the price of

silver?

Question 2
Consider the multivariate linear regression model that is described in Turkington (2007, p.71-
72.)
4
Business School
y=Xβ+u
The ordinary least squares estimator (OLS) of the vector of coefficients β is

̂ =(X’X)-1X’y
a) Must the matrix X’X be square? Must X be square?
b) The matrix X’X is invertible if X has full column rank (i.e. the columns of X are linearly
independent). What is the minimum number of rows (observations) if the model includes k
regressors and a constant?
c) For a given X, the predicted values of y are:

ŷ= X ̂ = X(X’X)-1X’y
Show that the matrix N = X(X’X)-1X’ is symmetric and idempotent. Is N invertible?
d) the prediction errors (residuals) are:

e = y -ŷ= y - X ̂ = y - X(X’X)-1X’y = y – Ny = (I-N)y = My

Show that the matrix M is symmetric and idempotent.

5
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 5. Quadratic Forms and Eigenvalues

Solve all questions. A selection of questions will be marked.


Question 1
Express the following quadratic forms as x’Ax where A is a symmetric matrix.

(a) 5 x1  3x2  4 x3
2 2 2

(b) 2 x1  9 x2  3x3  4 x1 x2  6 x1 x3  2 x2 x3
2 2 2

(c) x1  2 x3  5 x1 x3
2 2

Question 2
Consider the matrix
 2 2 1
1 
A  1 2 2 .
3 
 2 1 2 
Use the following two methods to show that A is an orthogonal matrix.
a) Check that the columns of A form an orthonormal set of vectors.
b) Check that A’ = A-1.

Question 3
Consider the multivariate linear regression model that is described in Turkington (2007, p.71-
72.)
y=Xβ+u
The ordinary least squares estimator (OLS) of the vector of coefficients β is

̂ =(X’X)-1X’y
The prediction errors (residuals) are:

e = y -ŷ= y - X ̂ = y - X(X’X)-1X’y = y – Ny = (I-N)y = My

Please represent the sum of squared prediction errors e’e as a quadratic form of the random
errors u.

6
Business School

Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 6. Topics in Matrix Algebra, Comparative Static Analysis in a Linear Model

Solve all questions. Questions with an asterisk (*) will be marked

Question 1
a) Show that the determinant of an orthogonal matrix is either 1 or -1.
b) Is an orthogonal matrix invertible (nonsingular). Why or why not?

Question 2*
a) Use the property of determinants |AB|=|A||B| to show that the determinant of an
idempotent matrix is either 0 or 1.
b) Show that the only idempotent matrix that is invertible (nonsingular) is the identity matrix.

Question 3
a) What does the covariance matrix of a multivariate probability distribution show?
b) Which of the following matrices is possibly a covariance matrix?
 5 3   5 1 4   2 5 4 2 4 5
       
A  0 7 B 1 3 2 C   5 2 3 D   9 7 2 
1 9   4 2 7   4 3 0   3 0 3 
    
Hint: Use R to calculate the eigenvalues where needed.
c) Are all covariance matrices nonsingular (invertible)? Explain.

Question 4*
Theorem: Cholesky Factorization
Any symmetric positive definite matrix A can be decomposed uniquely as
A = LL’
Where L is a lower triangular matrix with positive entries on the main diagonal.
Find the Choleski factorization for

 8 2 
A 
 2 5 
Question 5
The basic Keynesian macroeconomic model for a closed economy consists of the national
income identity which determines aggregate demand and a linear consumption function.
Y C  I G
C    Y
7
Business School
Using matrix notation:
Ax  b
 1  1  Y  I G
    =  
 1  C    
a) Show that the model has a unique solution.
b) Solve the model for vector of endogenous variables 𝑥, using matrix inversion.
c) Comparative static analysis: Suppose there is an investment boom in the mining sector that
increases 𝐼. Determine the response in the endogenous variables Y and C.

Question 6*
a)For the symmetric matrix

6 2
A =  
2 3
Find the orthogonal matrix Q such that Q’AQ is diagonal.
Hint: The columns of the orthogonal matrix Q are the eigenvectors of A.
b) What is the definiteness of A and the sign of x’Ax for all vectors x.

Question 7
Consider the following symmetric matrix A:
-4 2 0
 
A =  2 -5 2
 0 2 -2

Using the leading principal minors of A show that A is negative definite. What is the sign of
x’Ax for all x other than the null vector?

8
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 7. Differentiation

Solve all questions. Questions with an asterisk (*) will be marked.

Question 1
In each graph determine which function is the derivative of the other.
(A) (B)
f(x)

f(x)

g(x)

g(x)

Question 2
Verify the power rule of differentiation for y = x2.

Question 3*
Compute the slope of the following functions at the point x = 2. Apply the chain rule where
applicable.

5
(i) y= +7 (ii) y = (x3 - 7)5
x2
3
2
(iii) y = ln(x - 2) (iv) y = e x

Question 4*
a) Compute the gradient vector for the following functions:

(i) f ( x, y)  2 x y  5 xy  y
2 3 2 5

9
Business School
3
4
(ii) f ( x, y )  e2 xy

(iii) f ( x, y)  ln( x  3xy  5)


2

b) Compute the Hessian matrix for (i).

10
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 8. Implicit Function Theorem, Comparative Static Analysis in a Nonlinear


Model

Solve all questions. Questions with an asterisk (*) will be marked.

Question 1*
The generalized Cobb-Douglas production function is:
Y=AKαLβ 0 < α < 1, 0 < β < 1
a) Determine the marginal product of labor and its rate of change. Comment.
b) Is the Cobb-Douglas production function homogeneous? If so, of what degree?
c) Is the marginal product of labor a homogeneous function? Comment.
d) Is it possible that a production function has diminishing marginal products and at the
same time constant or even increasing returns to scale?

Question 2
Consider again the Cobb-Douglas production function given in question 2.
Show that the parameter β is the elasticity of output with respect to labor input. Derive the
elasticity both with and without logarithmic differentiation.

Question 3*
a) Does the nonlinear equation
x2 + 2y3 - 9z + 7 = 0
define an implicit function y = f(x, z) around the point y = 1, x = 3, z = 2?If so, find ∂y/∂x and
∂y/∂z and calculate these partial derivatives at the given point.
b) In what sense is the implicit function theorem about “compensatory “changes?

Question 4 (Chiang 2005,pp.205-207) *


Suppose that the demand for some good depends on its price and household income and the
supply depends only on price. Both functions may be nonlinear.
Demand: Sd = D(P, I)
Supply: Ss = S(P)
The price is the endogenous variable and income is exogenous, i.e. the price adjusts until the
market for the good is in equilibrium.
a) Market equilibrium requires:
F(P, I) = D(P, I) - S(P) = 0
What does the function F show? Under what condition does the implicit function P = f(I)
exist?Provide the necessary and sufficient condition.
b) Determine the effect of an exogenous increase in income on the price of the good. Assume

11
Business School
the good is normal. Illustrate with a graph.
Hint: Use the implicit function theorem to determine the comparative-static derivative
dP/dI.

Question 5
Derive the rate of technical substitution (RTS) for the Cobb-Douglas production.
Hint: Set the total differential equal to zero and calculate dK/dL.

Question 6

Are the following functions homogeneous? If so, of what degree?

(i) f ( x, y )  5 x y
2

x3
(ii) f ( x, y ) 
2 x 2 y  5 xy 2

(iii) f ( x, y )  xy  7
2

12
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 9. Taylor Approximation, Convexity, Optimization

Solve all questions. Questions with an asterisk (*) will be marked.

Question 1*
At x0 = 3, the value of the function y = x4 is 81. Now increase x by 0.1 and estimate the function
value, using Taylor’s approximation. How big is the error for a 1st order and 2st order
approximation? Illustrate with a graph.

Question 2
a) Show that the difference in the logarithm of a variable is approximately equal to the
difference measured in percent.
Hint: Use a first order Taylor’s approximation of the logarithmic function.
b) Show that In(1 + r) ≈ r for small r. Illustrate with a graph.
Hint: Calculate a first order Taylor’s approximation around the point 1 on the x axis.
This approximation is often used in financial economics. For an interest rate of 3% the
logarithm of the gross return (1 + r) is approximately equal to the interest rate r: ln(1 +
0.03) ≈ 0.03.

Question 3*
a) The function f(x1, …, xn) is strictly convex (concave) subset X of En if and only if the Hessian
matrix of f is positive(negative) definite at every point in X. Use this theorem to check out
the strict convexity or strict concavity of the following function:
f(x, y, z) = 3ex + 5y4 - lnz
b) For the following function find the critical points and classify them:
f(x, y) = y3 - x2 + 6x - 12y + 5

13
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 10. Optimization Problems

Solve all questions. Questions with an asterisk (*) will be marked.

Question 1*
Consider a consumer who buys x1 kilos of meat and x2 liters of petrol per month. The
consumer’s utility function is:
u(x1, x2) = ln(x1)+ ln(x2)
The domain of this function is the positive quadrant, R2 .

Suppose the price of petrol is $1.40 per liter and the price of meat is $28 per kilo. The total
amount that the consumer can spend on both goods is $560.
a) Formulate the consumer’s choice as a constrained optimization problem.
b) What are the first order conditions for this problem?
c) Show that the second order condition holds
d) How much petrol and meat will the consumer buy?
e) Show that the maximum you have obtained is unique (global maximum).

Question 2*
Suppose a firm that produces one good under perfect competition has a production function
Q(L, K) = L1/2 + K1/4
p is the price of output, w is the wage rate and r the cost of capital (rental).
a) Find an expression for total profit as a function of L and K.
b) Find the critical point of this function.
c) Show that this critical point is a unique global maximum.
Hint: Remember that the domain of the production function is defined for the first
quadrant R2 , i.e. L and K must be greater than zero.

d) How does the demand for labor change with a small change in w?

14
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 11. Comparative Static Analysis in Constrained Optimization Problems,


Envelope Theorem

Solve all questions. Questions with an asterisk (*) will be marked.

Question 1*
Use the utility function u(x1, x2) = lnx1 + lnx2 and the budget constraint p1x1 + p2x2 = I.
a) Find the demand functions. Also solve for the optimal Lagrangian multiplier, λ*.
b) Derive the indirect utility function (maximum value function).Differentiate the indirect
utility function with respect to income. Comment on result.
c) Use the envelope theorem to derive the result that you got in question (b).
d) Use the envelope theorem to derive the effect of an increase in p1 on consumer utility.

15
Business School
Economics Semester 2, 2025

Intermediate Mathematics for Economists (ECON2272)

Tutorial 12. Integration

Solve all questions. Questions will not be marked. A selection of questions will be discussed in
tutorial in Week 12.

Question 1
Find the indefinite integrals for:
i) ∫(7𝑥 + 5𝑥 − 4)d𝑥 ii) ∫𝑥 e d𝑥

Question 2
Evaluate the following definite integrals:

i) ∫ (2𝑥 − 4𝑥 + 5)d𝑥 ii) ∫ ( )d𝑥

Question 3
a) Use integration by substitution:
i) ∫ 4(𝑥 + 2) 𝑥 d𝑥 ii) ∫ 3𝑥 e d𝑥

b) Use integration by parts:


i) ∫ 𝑥 ln𝑥d𝑥 ii) ∫ ln𝑥d𝑥

Question 4
Evaluate the following expression at z = 5:

( )

What does this expression show? Illustrate with a graph.

16

Common questions

Powered by AI

Cramer's Rule solves systems of linear equations using determinants of the coefficient matrix and its modified counterparts for each variable. Each variable's solution is obtained by replacing the column of the matrix corresponding to the variable with the constant matrix and computing the determinant quotient. This method requires the determinant of the original matrix to be non-zero, thus only applicable to square and invertible matrices with unique solutions .

The implicit function theorem provides conditions for expressing a variable as a function of others in a nonlinear model, facilitating the analysis of how exogenous changes affect equilibria. It confirms the function's differentiability near an equilibrium point, aiding in understanding how market characteristics adjust to maintain equilibrium under varying parameters .

The covariance matrix in multivariate analysis encapsulates how changes in one variable affect others, measuring the degree to which variables change together. It must be symmetric and positive semidefinite to reflect valid variances and covariances. Additionally, the covariance matrix's eigenvalues must be non-negative to ensure validity .

In a general equilibrium model, an exogenous increase in demand for a joint product leads to increased supply and demand influences on its related products due to their complementary production. This results in price adjustments in both markets to reach a new equilibrium, taking into account the interconnectedness of their supply and demand curves .

The inverse of a matrix using elementary row operations involves augmenting the matrix with the identity matrix and performing row operations until the original matrix is transformed into the identity matrix, with the augmented part becoming the inverse. If successful, it signifies that the matrix is nonsingular or invertible .

Eigenvectors and eigenvalues are crucial in determining a matrix's definiteness. For symmetric matrices, if all eigenvalues are positive, the matrix is positive definite, and if all are negative, it is negative definite. The eigenvectors provide the directions of the principal axes, while the eigenvalues show the magnitude of stretch or compression along these axes, influencing the matrix's geometry and definiteness .

A linear system of equations has only the trivial solution when the determinant of its coefficient matrix is non-zero, meaning the matrix is invertible. For the system to have an infinite number of solutions, the determinant must be zero, indicating that the system is underdetermined or consistent with infinitely many solutions .

Taylor's theorem allows approximation of complex functions using polynomials, providing efficient solutions to nonlinear problems in economic models. It helps in local approximations by using function derivatives, capturing essential behavior around a point, while higher-order approximations offer greater precision, influencing decision-making and policy analysis .

An orthogonal matrix has vectors that are orthonormal for its columns, which implies that its determinant is either 1 or -1. This stems from the property that the determinant of an orthogonal matrix reflects its volume-preserving transformations, maintaining orthogonality and indicating invertibility .

Cholesky decomposition simplifies numerical linear algebra problems by breaking down a positive definite matrix into a product of a lower triangular matrix and its transpose. This decomposition aids in solving systems of linear equations, especially in efficient iterative algorithms, by reducing computational complexity and increasing numerical stability in symmetric, positive definite matrices .

You might also like