Proof: Showing C(n,0) < C(n,1) <...<C(n, floor(n/2)) = C(n, ceiling(n/2))

  • Thread starter Thread starter DrWillVKN
  • Start date Start date
  • Tags Tags
    Integer Positive
AI Thread Summary
The discussion focuses on proving the inequality C(n,0) < C(n,1) < ... < C(n, floor(n/2)) = C(n, ceiling(n/2)) > ... > C(n,n) for positive integers n. Participants suggest that induction may be a viable method, but express confusion about the flow of the proof and the use of combinatorial arguments. The relationship C(n, k) = C(n, n-k) is highlighted, emphasizing that the binomial coefficients reach their maximum at k = floor(n/2) and ceiling(n/2). There is also mention of using ratios, specifically C(n,k+1)/C(n,k), to establish the inequalities, but participants are uncertain about how to effectively apply this approach. Overall, the conversation reflects a struggle to clearly articulate the proof while considering different mathematical strategies.
DrWillVKN
Messages
20
Reaction score
0
Show that if n is a positive integer, then C(n,0) < C(n,1) <...<C(n, floor(n/2)) = C

Homework Statement


Show that if n is a positive integer, then C(n,0) < C(n,1) <...<C(n, floor(n/2)) = C(n, ceiling(n/2)) > ... > C(n, n)

Homework Equations



none

The Attempt at a Solution



Seems pretty direct, but I could be wrong. I suppose induction could work, too (or might be necessary).

The statement seems to imply that C(n, floor(n/2)) > C(n, k) for floor(n/2) > k >= 0 and C(n, ceiling(n/2)) > C(n, k) for ceiling(n/2) > k >= 0; n and k are positive integers. This can be stated because C(n, k) = C(n, n-k). P(n, k) = C(n, k)*k! by basic counting observation. Do these parts still need to be proven before they are shown to be equal? Does one need to use pascal's identity for this?

The trouble I have with the problem is with the flow of this proof. C(n, floor(n/2)) > C(n, k) for n>= k >= 0 because C(n, k) is greatest when k = n/2 for even, and k = (n+1)/2 = ceiling(n/2) and k = (n-1)/2 = floor(n/2) for odd, yet I don't know how I would go about stating this. It's too basic and intuitive.

Since C(n, k) = P(n, k)/k!, there are two cases to show:

1) When k is even, C(k, floor(k/2)) = k!/[(k/2)!(k/2)!], and C(k, ceiling(k/2)) = n!/[(k/2)!(k/2)!].
2) When k is odd, C(k, floor(k/2)) = k!/[(k-1/2)!(k+1/2)!] and C(k, ceiling(k/2)) = k!/[(k+1/2)!(k-1/2)!].

...

Thus, C(n, floor(n/2)) = C(n, ceiling(n/2)) > C(n, k) for n >=k >= 0 .
 
Last edited:
Physics news on Phys.org


I would try to write down an expression for the ratio C(n,k+1)/C(n,k). If it's greater than one C(n,k+1)>C(n,k), if it's less than one C(n,k+1)<C(n,k).
 


So the ratio would fit everything on the left and right side in place, and then it comes together when k hits the floor and ceiling, and they're joined by equating them? I'm still confused about how to use the ratio (and for combinatorial proofs in general).

I have a hard time doing it by combinatorial proof, but by induction I have (for inductive step) the inductive hypothesis being the statement, adding the k + 1 term to the k term and showing pascal's identity, and then joining the left and right hand side by showing the ceiling and floor sets in the inductive step to be equal using algebra
 
Last edited:


DrWillVKN said:
So the ratio would fit everything on the left and right side in place, and then it comes together when k hits the floor and ceiling, and they're joined by equating them? I'm still confused about how to use the ratio (and for combinatorial proofs in general).

I have a hard time doing it by combinatorial proof, but by induction I have (for inductive step) the inductive hypothesis being the statement, adding the k + 1 term to the k term and showing pascal's identity, and then joining the left and right hand side by showing the ceiling and floor sets in the inductive step to be equal using algebra.

I don't think induction is the easy way to do it. Did you figure out a simple expression for C(n,k+1)/C(n,k)? Just looking at it might help you organize your thoughts.
 
I tried to combine those 2 formulas but it didn't work. I tried using another case where there are 2 red balls and 2 blue balls only so when combining the formula I got ##\frac{(4-1)!}{2!2!}=\frac{3}{2}## which does not make sense. Is there any formula to calculate cyclic permutation of identical objects or I have to do it by listing all the possibilities? Thanks
Essentially I just have this problem that I'm stuck on, on a sheet about complex numbers: Show that, for ##|r|<1,## $$1+r\cos(x)+r^2\cos(2x)+r^3\cos(3x)...=\frac{1-r\cos(x)}{1-2r\cos(x)+r^2}$$ My first thought was to express it as a geometric series, where the real part of the sum of the series would be the series you see above: $$1+re^{ix}+r^2e^{2ix}+r^3e^{3ix}...$$ The sum of this series is just: $$\frac{(re^{ix})^n-1}{re^{ix} - 1}$$ I'm having some trouble trying to figure out what to...
Back
Top