Proving Convexity: Steps and Conditions for a Set to be Convex

  • MHB
  • Thread starter rputra
  • Start date
In summary: We used that $0\le k\le 1$ and hence $0\le 7k\le 7$.Instead of $(1-k)(x_1+y_1) \leq (x_1+y_1)$ try $(1-k)(x_1+y_1) \leq 7(1-k)$ and $(1-k)(x_1+y_1) \leq 7(1-k)$ and similarly for $k(x_2+y_2)$. We used that $0\le k\le 1$ and hence $0\le 7k\le 7$.
  • #1
rputra
35
0
I would like getting for this problem: Consider $\mathscr D := \{(x,y) \mid x, y \in \mathbb R^2 \}$ with $x+y \geqslant 0$, $x+y \leq 7$ and $x \geqslant 2$. Show that the set is convex. The standard steps say that there exist $k_1, k_2 \geqslant 0$ with $k_1 + k_2 = 1$, and I have to prove that $xk_1 + y(1-k_1) \in \mathscr D$ in order to show convexity. Please help me on going forward from here, thank you very much for your time and effort.
 
Physics news on Phys.org
  • #2
Tarrant said:
I would like getting for this problem: Consider $\mathscr D := \{(x,y) \mid x, y \in \mathbb R^2 \}$ with $x+y \geqslant 0$, $x+y \leq 7$ and $x \geqslant 2$.
This definition should say
\[
\mathscr D=\{(x,y)\in\mathbb{R}^2\mid x+y \geqslant 0,x+y \leq 7,x \geqslant 2\}.
\]

Tarrant said:
Show that the set is convex.
Note that $\mathscr D=D_1\cap D_2\cap D_3$ where $D_1=\{(x,y)\mid x+y \geqslant 0\}$, $D_2=\{(x,y)\mid x+y \leq 7\}$ and $D_3=\{(x,y)\mid x \geqslant 2\}$. I suggest proving a general fact that that intersection of convex sets is convex, so it would be sufficient to prove that each of $D_i$ is convex.

Tarrant said:
The standard steps say that there exist $k_1, k_2 \geqslant 0$ with $k_1 + k_2 = 1$, and I have to prove that $xk_1 + y(1-k_1) \in \mathscr D$ in order to show convexity.
Yes, points lying on the segment between $(x_1,y_2)$ and $(x_2,y_2)$ have coordinates $((1-k)x_1+kx_2,(1-k)y_1+ky_2)$ for $0\le k\le 1$. Use this definition to to prove that sets $D_i$ are convex. For example, for $D_2$ you need to show that $x_1+y_1\le 7$ and $x_2+y_2\le 7$ imply
\[
(1-k)x_1+kx_2+(1-k)y_1+ky_2\le7.
\]
 
Last edited:
  • #3
Evgeny.Makarov said:
Yes, points lying on the segment between $(x_1,y_2)$ and $(x_2,y_2)$ have coordinates $((1-k)x_1+kx_2,(1-k)y_2+ky_2)$ for $0\le k\le 1$. Use this definition to to prove that sets $D_i$ are convex. For example, for $D_2$ you need to show that $x_1+y_1\le 7$ and $x_2+y_2\le 7$ imply
\[
(1-k)x_1+kx_2+(1-k)y_2+ky_2\le7.
\]

Thank you for your quick response! I understand the first part and second part of your solution, but on the third part (quoted above,) how do you get from $x_1+y_1\le 7$ and $x_2+y_2\le 7$ to implying that $(1-k)x_1+kx_2+(1-k)y_2+ky_2\le7$? I would appreciate if you could provide me with additional steps. Thank you again and again for your time and effort.

PS: I am sorry I get back late answering your question.
 
  • #4
Tarrant said:
how do you get from $x_1+y_1\le 7$ and $x_2+y_2\le 7$ to implying that $(1-k)x_1+kx_2+(1-k)y_2+ky_2\le7$?
There was a typo in the coordinates of a point on the segment and in the inequality to be proved. The point has coordinates $((1-k)x_1+kx_2, (1-k)y_1+ky_2)$ ($y_1$ instead of $y_2$), and the claim should read
\[
(1-k)x_1+kx_2+(1-k)y_1+ky_2\le7.
\]
Note that $(1-k)x_1+kx_2+(1-k)y_1+ky_2=(1-k)(x_1+y_1)+k(x_2+y_2)$. Now use conditions on $(x_1,y_1)$ and $(x_2,y_2)$ and the fact that $0\le k\le 1$ (it is essential).
 
Last edited:
  • #5
Evgeny.Makarov said:
There was a typo in the coordinates of a point on the segment and in the inequality to be proved. The point has coordinates $((1-k)x_1+kx_2, (1-k)y_1+ky_2)$ ($y_1$ instead of $y_2$), and the claim should read
\[
(1-k)x_1+kx_2+(1-k)y_1+ky_2\le7.
\]
Note that $(1-k)x_1+kx_2+(1-k)y_1+ky_2\le7=(1-k)(x_1+y_1)+k(x_2+y_2)$. Now use conditions on $(x_1,y_1)$ and $(x_2,y_2)$ and the fact that $0\le k\le 1$ (it is essential).

Thank you again. Let me think about it and get back with you soon. Thanks again.
 
  • #6
Evgeny.Makarov said:
...
Note that $(1-k)x_1+kx_2+(1-k)y_1+ky_2\le7=(1-k)(x_1+y_1)+k(x_2+y_2)$. Now use conditions on $(x_1,y_1)$ and $(x_2,y_2)$ and the fact that $0\le k\le 1$ (it is essential).

I know easily that because $0 \leq k \leq 1$ and $0 \leq (1-k) \leq 1$ therefore $(1-k)(x_1+y_1) \leq (x_1+y_1) \leq 7$ and $k(x_2+y_2) \leq (x_2+y_2) \leq 7$, hence $(1-k)(x_1+y_1) \leq 7$ and $k(x_2+y_2) \leq 7$. And then, what is next? How do you go from here to get to the claim $(1-k)x_1+kx_2+(1-k)y_1+ky_2\leq 7$? I am sorry but I am still confused. Thanks again in advance for all your time and help.s
 
  • #7
Tarrant said:
I know easily that because $0 \leq k \leq 1$ and $0 \leq (1-k) \leq 1$ therefore $(1-k)(x_1+y_1) \leq (x_1+y_1) \leq 7$
Instead of $(1-k)(x_1+y_1) \leq (x_1+y_1)$ try $(1-k)(x_1+y_1) \leq 7(1-k)$ and similarly for $k(x_2+y_2)$.
 

FAQ: Proving Convexity: Steps and Conditions for a Set to be Convex

1. What is convexity and why is it important?

Convexity is a mathematical property of a set that describes its shape and structure. A set is convex if any line segment connecting two points in the set lies entirely within the set. Convexity is important because it has many applications in optimization, economics, and geometry.

2. What are the steps to prove convexity of a set?

The steps to prove convexity of a set are as follows:

  1. Choose any two points in the set.
  2. Formulate the line segment connecting these two points using a parameter t.
  3. Verify that this line segment lies entirely within the set, by showing that all points on the line segment satisfy the conditions of the set.
  4. If the line segment is within the set for all possible values of t, then the set is convex.

3. What are the necessary conditions for a set to be convex?

The necessary conditions for a set to be convex are:

  1. The set must be closed. This means that it contains all its boundary points.
  2. The set must be convex in the strict sense. This means that the line segment connecting any two points in the set must lie entirely within the set, and not just on the boundary.

4. Can a set be convex in one dimension but not in higher dimensions?

Yes, a set can be convex in one dimension but not in higher dimensions. For example, a line segment is convex in one dimension, but a curved line or a surface in higher dimensions may not be convex.

5. Are there any practical applications of convexity?

Yes, convexity has many practical applications, including:

  1. In optimization, convexity is used to find the minimum or maximum values of a function.
  2. In economics, convexity is used to model consumer preferences and production costs.
  3. In geometry, convexity is used to study the properties of shapes and their transformations.

Similar threads

Replies
10
Views
2K
Replies
4
Views
708
Replies
1
Views
1K
Replies
1
Views
1K
Replies
52
Views
3K
Replies
4
Views
2K
Back
Top