Proving the Existence of a Subset with Divisible Sum in Number Theory

Click For Summary

Homework Help Overview

The problem involves proving that for a natural number n and a subset A of natural numbers with n elements, there exists a subset of A such that the sum of its elements is divisible by n. This falls under the subject area of number theory.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • The original poster attempts to use the pigeonhole principle to find a solution but expresses difficulty in proving the existence of the subset. Some participants suggest considering the number of congruency classes modulo n and explore implications of sums of elements in relation to these classes.

Discussion Status

Participants are actively engaging with the problem, exploring various approaches and raising questions about congruency classes and modular arithmetic. Hints have been offered to guide the original poster's reasoning without providing direct solutions.

Contextual Notes

The discussion reflects a lack of explicit consensus on the approach, with participants considering different aspects of the problem and the implications of their reasoning.

AdrianZ
Messages
318
Reaction score
0

Homework Statement


Suppose n is a natural number and A is a subset of natural number with n elements. Prove that a subset of A exists that the sum of its elements is dividable by n.

The Attempt at a Solution


well, This problem is harder than I can solve it. I first tried to use the pigeonhole principle to prove the existence of such a subset of A. well, It wasn't easy to prove it that way as I thought. so, now I feel that I'm stuck and have no idea of how to solve it. Any ideas?
 
Physics news on Phys.org
You're on the right track with PHP. How many congruancy classes are there modulo n? Think about this for a bit, and I have another hint if you need one.
 
Robert1986 said:
You're on the right track with PHP. How many congruancy classes are there modulo n? Think about this for a bit, and I have another hint if you need one.

I've already thought about the number of congruency classes a lot. There are n congruency classes of modulo n ([0],...,[n-1]). In fact the Euclidean algorithm of division forces that. but the question is still so confusing. so another hint would be appreciated.
 
OK, let a_1,a_2,\ldots,a_n be the elements of your n-element set. Consider these sums:

a_1 \equiv m_1 mod n
a_1 + a_2 \equiv m_2 mod n
\vdots
a_1 + a_2 + \cdots + a_n \equiv m_n mod n

Now, what would it mean if one of the m_i's were 0? OK, now what if none of them are 0? How many congruancy classes does this leave? How many sums are there above? If a,b,c are integers and a \equiv c mod n and b \equiv c mod n what can you say about a-b mod n?
 
Robert1986 said:
OK, let a_1,a_2,\ldots,a_n be the elements of your n-element set. Consider these sums:

a_1 \equiv m_1 mod n
a_1 + a_2 \equiv m_2 mod n
\vdots
a_1 + a_2 + \cdots + a_n \equiv m_n mod n

OK, now what if none of them are 0? How many congruancy classes does this leave?How many sums are there above? If a,b,c are integers and a \equiv c mod n and b \equiv c mod n what can you say about a-b mod n?

well, if one of them is zero, we're done. if none of them is zero, then we'll have n-1 congruency classes and n such sums. that would mean that two of these sums should be in the same congruency class. therefore if we subtract those two from each other we'll have a new sum that is congruent to 0 modulo n and we're done.

Thanks for the help.
 
Beautiful!
 

Similar threads

Replies
1
Views
2K
Replies
20
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 8 ·
Replies
8
Views
11K
  • · Replies 2 ·
Replies
2
Views
5K
  • · Replies 18 ·
Replies
18
Views
3K
Replies
2
Views
1K
  • · Replies 10 ·
Replies
10
Views
2K