0% found this document useful (0 votes)
2 views2 pages

Problem Set 1 Solutions

The document presents solutions to a problem set in ECO 503, discussing relations between countries and their capitals, properties of preference relations, and the implications of convexity and quasi-concavity in economic theory. It includes mathematical proofs demonstrating the uniqueness of capital assignments, transitive properties of relations, and the relationship between convex preferences and quasi-concavity. The solutions are structured in a formal mathematical style, addressing various scenarios and assumptions related to the defined relations.

Uploaded by

zsiam14
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)
2 views2 pages

Problem Set 1 Solutions

The document presents solutions to a problem set in ECO 503, discussing relations between countries and their capitals, properties of preference relations, and the implications of convexity and quasi-concavity in economic theory. It includes mathematical proofs demonstrating the uniqueness of capital assignments, transitive properties of relations, and the relationship between convex preferences and quasi-concavity. The solutions are structured in a formal mathematical style, addressing various scenarios and assumptions related to the defined relations.

Uploaded by

zsiam14
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

ECO 503 Problem Set 1 Solutions

1. Let W be the set of all countries, and let C be the set of all cities. Define the relation R from W to C as R =

{(x, y) → W ↑ C : x’s capital is y}. Now, let

S = {Benin, Bolivia, Burundi, Eswaitini, Indonesia, Ivory Coast, Malaysia, Netherlands, South Africa, Sri Lanka, Afghanistan},

and let S = S ↓ {Nauru}.


Recall that the in order for the described relation R to be a function from some set W to C, we need all countries in W
→ →

to have a unique capital city in C. In the set of all real world countries W , there are some countries with more than one

capital city (such as South Africa), and there is also the country Nauru which has no capital city. Both of these prevent

the given relation R from being a function. If we exclude the problematic countries to end up with a new set of countries
! "
W = W \ S , and define a similar relation R from W to C as R = (x, y) → W ↑ C : x’s capital is y , then R assigns
→ → → → → → →

a unique capital city to every country in W , and hence R : W ↔ C is then a function.


→ → →

2. Here, we assume ↭ are complete and transitive. We have to show that:

(a) If x ↭ y and y ↗ z, then x ↗ z.

Suppose x ↭ y and y ↗ z. y ↗ z means y ↭ z and ¬ (z ↭ y). Because we have x ↭ y and y ↭ z, and ↭ is transitive,

therefore we can say x ↭ z. Towards a contradiction, assume z ↭ x. Now, z ↭ x and we also had x ↭ y, so by

transitivity of ↭, we then have z ↭ y, which is a contradiction because we also have ¬ (z ↭ y). So ¬ (z ↭ x), combining

this with x ↭ z, we get x ↗ z.

(b) If x ↗ y and y ↭ z, then x ↗ z.

Suppose x ↗ y and y ↭ z. That means x ↭ y and ¬ (y ↭ x). By transitivity, we have x ↭ z. Towards a contradiction,

assume z ↭ x. Because y ↭ z and z ↭ x, by transitivity we get y ↭ x, which is a contradiction. Therefore, we have

x ↭ z and ¬ (z ↭ x), which means x ↗ z.

3. Suppose ↗ is asymmetric and negatively transitive.

First, we prove that ↘ is reflexive. Take arbitrary x → X . We have to show x ↘ x. By definition of ↘, this means showing

¬ (x ↗ x). Towards a contradiction, assume x ↗ x, because ↗ is asymmetric, this means ¬ (x ↗ x). So now we have both

x ↗ x and ¬ (x ↗ x), an obvious contradiction. Therefore, ¬ (x ↗ x), which means x ↘ x.

Next, we prove that ↘ is symmetric. Take arbitrary x, y → X . Suppose x ↘ y. We have to show y ↘ x. Because x ↘ y,

by definition we have ¬ (x ↗ y) and ¬ (y ↗ x). Writing in a di!erent order, we have ¬ (y ↗ x) and ¬ (x ↗ y). This, by

1
definition of ↘, means y ↘ x.

Last, we prove that ↘ is transitive. Take arbitrary x, y, z → X , and assume x ↘ y and y ↘ z. We have to show x ↘ z.

Because x ↘ y, we can say ¬ (x ↗ y). Because y ↘ z, we can say ¬ (y ↗ z). Because ↗ is negatively transitive, we have

¬ (x ↗ z).

Because y ↘ z, we have ¬ (z ↗ y). Because x ↘ y, we have ¬ (y ↗ x). By negative transitivity of ↗, we can say ¬ (z ↗ x).

Now we have both ¬ (x ↗ z) and ¬ (z ↗ x). Together, they mean x ↘ z by definition.

4. We take rational preferences ↭.

(a) Suppose ↭ are convex. For any z → X , let x, y → Bz be arbitrary. This means x ↭ z and y ↭ z. For arbitrary

ω → [0, 1], consider the convex combination ωx + (1 ≃ ω) y.

Now, without loss of generality, we can assume x ↭ y and proceed with the proof, because the arguments would be

analogous for y ↭ x. Because ↭ are convex, ωx + (1 ≃ ω) y ↭ y, and we also have y ↭ z. By transitivity of ↭, we can

say ωx + (1 ≃ ω) y ↭ z, hence ωx + (1 ≃ ω) y → Bz . This shows that Bz is a convex set.

(b) Now suppose for any z → X , Bz is a convex set. Let x, y → X be arbitrary. Because ↭ are complete, either x ↭ y or

y ↭ x. WLOG, assume x ↭ y, and we have y ↭ y. So x, y → By , and so for any ω → [0, 1], ωx + (1 ≃ ω) y → By , which

means ωx + (1 ≃ ω) y ↭ y. This shows that ↭ are convex.

5. ( ⇐= ) Suppose u is quasi-concave, and WLOG1 , assume x ↭ y, ↫ u (x) ⇒ u (y) and min {u (x) , u (y)} = u (y). Because

u is quasi-concave, u (ωx + (1 ≃ ω) y) ⇒ min {u (x) , u (y)} = u (y), and so ωx + (1 ≃ ω) y ↭ y, which shows that ↭ are

convex.

( =⇑ ) Now suppose ↭ are convex, and let x, y → X be arbitrary. Because ↭ are complete, either x ↭ y, or y ↭ x.

Case 1. x ↭ y. Then, u (x) ⇒ u (y), and min {u (x) , u (y)} = u (y). Because ↭ are convex, ωx + (1 ≃ ω) y ↭ y, so

u (ωx + (1 ≃ ω) y) ⇒ u (y) = min {u (x) , u (y)}.

Case 2. y ↭ x. Then, u (y) ⇒ u (x), and min {u (x) , u (y)} = u (x). Because ↭ are convex, ωx + (1 ≃ ω) y ↭ x, so

u (ωx + (1 ≃ ω) y) ⇒ u (x) = min {u (x) , u (y)}.

So in either case, we have u (ωx + (1 ≃ ω) y) ⇒ min {u (x) , u (y)}, which means u is quasi-concave.

1 [Link]

You might also like