Maximizing Sets: A Problem with Sets

  • Thread starter Thread starter dapet
  • Start date Start date
  • Tags Tags
    Sets
Click For Summary
The discussion centers on maximizing the number of two-element sets X_i that share at least one common element with a chosen set C, derived from two sets A and B. Participants explore the conditions under which the constant c can be determined, specifically aiming to establish that c can be at most 3/4. There is confusion regarding the definition of set C and the examples provided, particularly how elements are selected from sets A and B. Clarifications are sought on the construction of sets X_i and the implications of the example given. The thread highlights the complexity of the problem and the need for a clearer understanding of the parameters involved.
dapet
Messages
9
Reaction score
0
Let's imagine two sets A = {1,2,...,k} and B = {-1,-2,...,-k} for some natural k, then let's create n two-element sets X_1,X_2,...,X_n such that for each 0<i=<n X_i = {a,b} where a is from A and b is from B but |a|<>|b|. We know how do sets X_i look like and according to this we will choose the set C = {c_1,c_2,...,c_k} where |c_i|=i such that the number (denoted MAX) of sets X_1,X_2,...,X_n that have at least one common element with C is maximal. Determine the maximal constant 0<=c<=1 such that MAX>=[cn] for arbitrary n,k and sets X_1,X_2,...,X_n.
NOTE: [x] denotes the integral part of number x

Example:
k = 2, A = {1,2}, B = {-1,-2}
n = 4, X_1 = {-1,-2}, X_2 = {-1,2}, X_3 = {1,-2}, X_4 = {1,2}
we can choose C = {1,2} (in this case we have more possibilities) the number X_i that have at least one common element with C is 3, X_1 and C have no common element.

From this example we can easily see, that c<=3/4, I think that c=3/4 is sufficient condition, but I can't prove it.
Could anybody help me with it? Thanks.
 
Physics news on Phys.org
This doesn't make a whole lot of sense. For one:
we will choose the set C = {c_1,c_2,...,c_k} where |c_i|=i
To me, this says that C = {1, 2, 3, 4, ..., k}. It seems that the set C will always look like this. From there, it is still not clear as to exactly what we're looking for. It is further confused by your example because you said:
for each 0<i=<n X_i = {a,b} where a is from A and b is from B
But then gave, as an example:
X_4 = {1,2}
which takes both its elements from A.
 
The standard _A " operator" maps a Null Hypothesis Ho into a decision set { Do not reject:=1 and reject :=0}. In this sense ( HA)_A , makes no sense. Since H0, HA aren't exhaustive, can we find an alternative operator, _A' , so that ( H_A)_A' makes sense? Isn't Pearson Neyman related to this? Hope I'm making sense. Edit: I was motivated by a superficial similarity of the idea with double transposition of matrices M, with ## (M^{T})^{T}=M##, and just wanted to see if it made sense to talk...

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 6 ·
Replies
6
Views
1K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 20 ·
Replies
20
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 25 ·
Replies
25
Views
3K