UNIVERSITY OF DELHI

M.A. ECONOMICS SUMMER SEMESTER

COURSE 002. INTRODUCTORY MATHEMATICAL ECONOMICS
Solution to Final Examination
30th November 2012

PART A
1.

A primitive farmer seeks to maximize the present value of her utility from the production
of apples. Let xt be the number of apples at time t, and zt is the rate at which she eats

the apples, then change in the stock of apples at time t is expressed:

x = abxt zt .
Her objective is to maximize the present value of utility, which she gets from two
sources, the sale of apples at the price p, which occurs at the end of the season, T , and
her personal consumption of apples during the season, u(zt ) = ln(zt ). The constraints
xt > 0 and zt > 0 can be assumed to hold (i.e., you do not need to explicitly write these
constraints).
(a) Formally state the decision makers optimization problem.
ANS:

maxzt

RT
0

ert ln(zt )dt + erT pxT

x = a bxt zt
(b) Specify the Hamiltonian (current or present value) and the first-order necessary
conditions for an optimum.
subject to

ANS:

Hc
Hc
zt

= ln(zt ) + (a bxt zt )
1
zt

=0

Hc
xt

= t b = rt t

Hc
t

=a bxt

zt =x

and TC: T = p

(c) Give an economic interpretation of the each of the first-order conditions and the
transversality condition.
ANS: Dorfman-like interpretation expected. For TC, the marginal value of the apples must equal the
price at that point.

(d) Solve for the optimal path for consumption of the apples. Be sure to express your
solution in terms of the known parameters of the model, a, b, p, and T .
(3, 4, 4, 5)
ANS:

t b = rt t

=> t = (r + b)t

=> A = pe(r+b)t

=> t = Ae(r+b)t = p

From TC: T = p

=> t = Ae(r+b)t

So t = pe(r+b)t e(r+b)t = pe(r+b)(tT )

From Hz = 0 we get 1
zt

=> zt = p1 e(r+b)(T t)

= t = pe(r+b)(tT )

PART B
2. (a)

Let S be and invertible n n matrix and let A be an n m matrix. Show that

the dimension of the kernel of A is equal to that of the kernel of SA and that the
dimension of the image of A is equal to that of the image of SA.
ANS: Consider the kernel of SA. It is in fact the same as the kernel of A, because

ker(SA) = { x <m |SA x = 0 } = { x <m |A x = S 1 0 }

= { x <m |A x = 0 } = ker(A)
1

Therefore the dimension of the kernel of SA is the same as that of the ker(A).
Combining this fact with the rank-nullity theorem gives the following sequence of equation:
dim(image(A)) = m dim(ker(A)) = m dim(ker(SA)) = dim(image(SA)).
(b)

Prove that if P L(V ) is such that P 2 = P and every vector in null P is orthogonal
to every vector in range P , then P is an orthogonal projection..
ANS: Supposee P L(V ) is such that P 2 = P every vector in ker P is orthogonal to every
vector in range P. Let U=range P. We will show P equals to the orthogonal projection PU . To do
that , suppose v V. Then v = P v + (v P v) .
Clearly, Pv rangeP = U . Also P (v Pv ) = P v P 2 v = 0 which means that v P v ker P.
Thus v P v is orthogonal to every vector in U. In other words (v P v) U . Thus the equation
above writes v as the sum of a vector in U and U . In this decomposition, the vector U equals
PU v. Hence P v = PU v, as desired.

(c)

For every T L(V, W ) prove the followings (i)

dim null T = dim null T + dimW dimV
ANS: Let T L(V, W ). Then -

and

dim ker T = dim(rangeT ) = dim W dim(rangeT ) = dim ker T dim W dim V

where first and equalities are directly coming from the rank-nullity theorem, the second equality
can be done as follows as part of next question -

(ii)
ANS:
3.

dim range T = dim range T .

(4, 4, 4)

dim(rangeT ) = dim W dim nullT = dim V dim nullT = dim rangeT

Consider the subspace V = {

x <3 |2x1 + 6x2 4x3 = 0}.
(a)

Find a basis for V. What is its dimension of this basis?

Notice that V is the kernel of the linear transformation given by the matrix [2 6 -4]. The

parametrized solution to the equation [ 2 6 -4]

x = 0 is

3
3t + 2s
<3 |t < . Thus, 1 , 0 is a basis for V. V is two dimensional,

1
0
s
because its basis consists of two vectors.

(b)

Extend the basis from part (a) to a basis for <3 .

1
2
3

1 , 0 , 3 is a basis

0
1
2

(c)

<3 .

Write the matrix of the linear transformation T : <3 < such that T
x =
x .[.],
where [.] is the third vector in part (b).
(2, 2, 2)
[0 0 14] is the desired matrix in part (b). Note this depends on the third vector you choose there.

PART C
4.

Solve the following problem.

Minimize e(x1 x2 ) s.t.
ex1 + ex2 20
0 x1 .

5.

(18)

Define a function f : <n < to be coercive if for every sequence (xk ) in <n s.t.
||xk || , we have f (xk ) . Now suppose f is coercive and continuous. Show
that then, for all <, the lower contour set LCS(a) {x|f (x) } is compact.
(18)

6.

Consider the following system of differential equations.

Dx1 = (A Ax2 Bx1 )x1
Dx2 = (Cx1 C)x2
where A, B > 0, A > B, C > 1, and AC > B. Dx1 and Dx2 are time derivatives.
Interpret the equations as a predator-prey model. How does it differ from the
Lotka-Volterra model. Be very brief.
(4)
(b) Find all equilibria in which x1 > 0. Using the linearization theorem, comment on
the stability of these equilibria.
(14)
(a)