Proving Combinations Formula | Step-by-Step Guide and Examples

  • Thread starter Thread starter Government$
  • Start date Start date
  • Tags Tags
    Combinations Proof
AI Thread Summary
The discussion revolves around proving the combinations formula, specifically the identity involving binomial coefficients: \(\binom{n}{m}=\binom{n}{m-2}+2\binom{n-1}{m-1}+\binom{n-2}{m-2}\). Participants express confusion about the steps needed to manipulate the formula and check for possible typos in the original problem statement. Several attempts to simplify the expressions and find a common denominator are shared, highlighting the complexity of the algebra involved. Ultimately, there is a suggestion that the problem may have been misstated in the textbook, leading to further exploration of related identities. The conversation emphasizes the challenges of proving combinatorial identities and the importance of clarity in mathematical expressions.
Government$
Messages
87
Reaction score
1

Homework Statement


Prove the following:

\frac{n!}{m!(n-m)!} = \frac{n!}{(m-2)!(n-m + 2)!} + 2* \frac{(n-1)!}{(m-1)!(n-m)!} + \frac{(n-2)!}{(m-2)!(n-m)!}



The Attempt at a Solution



I tried writing following

\frac{n!}{m!(n-m)!} = \frac{n!}{m!(n-m)!}(\frac{m(m-1)}{(n-m + 2)(n-m + 1)} + \frac{2m}{n} + \frac{m(m-1)}{n(n-1)})

Now i should get \frac{m(m-1)}{(n-m + 2)(n-m + 1)} + \frac{2m}{n} + \frac{m(m-1)}{n(n-1)} = 1
but i don't get 1. Any other ideas?
 
Physics news on Phys.org
Hint: m! = m * (m-1) * (m-2)!
 
I know that, but where to use it?
 
your m(m-1)/n(n-1) is wrong :redface:
 
I rechecked it two or three time i really don't see how it is wrong.
 
Government$ said:

Homework Statement


Prove the following:

\frac{n!}{m!(n-m)!} = \frac{n!}{(m-2)!(n-m + 2)!} + 2* \frac{(n-1)!}{(m-1)!(n-m)!} + \frac{(n-2)!}{(m-2)!(n-m)!}
What is the lowest common denominator for the right-hand side ?

The Attempt at a Solution



I tried writing following

\frac{n!}{m!(n-m)!} = \frac{n!}{m!(n-m)!}(\frac{m(m-1)}{(n-m + 2)(n-m + 1)} + \frac{2m}{n} + \frac{m(m-1)}{n(n-1)})

Now i should get \frac{m(m-1)}{(n-m + 2)(n-m + 1)} + \frac{2m}{n} + \frac{m(m-1)}{n(n-1)} = 1
but i don't get 1. Any other ideas?
 
(m-1)!(n-m+2)!
 
I don't understad why, since (m-2)! doesn't contain (m-1) nor (n-m)! contains (n-m+2)! or (n-m+1)!
 
If i want get whole expression under one fraction i should multply first fraction with (m-1)/(m-1) to get n!(m-1)/(m-1)!(n-m+2)! , second fraction with (n-m+2)(n-m+1)/(n-m+2)(n-m+1) to get
(n-1)!(n-m+2)(n-m+1)/(m-1)!(n-m+2)! , and third with (m-1)(n-m+2)(n-m+1)/(m-1)(n-m+2)(n-m+1)
to get (n-2)!(m-1)(n-m+2)(n-m+1)/(m-1)!(n-m+2)!
 
  • #10
Government$ said:
I don't understad why, since (m-2)! doesn't contain (m-1) nor (n-m)! contains (n-m+2)! or (n-m+1)!
You were correct. I deleted my post, but you must have seen it before I deleted it.
 
  • #11
Government$ said:
If i want get whole expression under one fraction i should multply first fraction with (m-1)/(m-1) to get n!(m-1)/(m-1)!(n-m+2)! , second fraction with (n-m+2)(n-m+1)/(n-m+2)(n-m+1) to get
(n-1)!(n-m+2)(n-m+1)/(m-1)!(n-m+2)! , and third with (m-1)(n-m+2)(n-m+1)/(m-1)(n-m+2)(n-m+1)
to get (n-2)!(m-1)(n-m+2)(n-m+1)/(m-1)!(n-m+2)!
That's virtually unreadable as it is written. Break it up into separate lines & give it some space...

As follows:
If i want get whole expression under one fraction i should multiply:

first fraction with (m-1)/(m-1) to get
n!(m-1)/(m-1)!(n-m+2)! ,​
second fraction with (n-m+2)(n-m+1)/(n-m+2)(n-m+1) to get
(n-1)!(n-m+2)(n-m+1)/(m-1)!(n-m+2)! ,​
and third with (m-1)(n-m+2)(n-m+1)/(m-1)(n-m+2)(n-m+1)
to get
(n-2)!(m-1)(n-m+2)(n-m+1)/(m-1)!(n-m+2)!​

So the numerator is (m-1)n! + 2(n-m+2)(n-m+1)(n-1)! + (m-1)(n-m+2)(n-m+1)(n-2)! .

There's a common factor of (n-2)! , right?
 
  • #12
After that i get (n-2)!*(n(n-1)(m-1) + 2(n-m+2)(n-m+1)(n-1) + (m-1)(n-m+2)(n-m+1))/(m-1)!(n-m+2)!

i tried and i can't find any other common factor in numerator and i have common factors between two products such as (n-1) , (m-1), (n-m+2)(n-m+1). Only other way is to multiply everything and then try to factor out something that will cancel m-1 and (n-m+2)(n-m+1) in denominator.
 
  • #13
Government$ said:
\frac{n!}{m!(n-m)!} = \frac{n!}{(m-2)!(n-m + 2)!} + 2* \frac{(n-1)!}{(m-1)!(n-m)!} + \frac{(n-2)!}{(m-2)!(n-m)!}
Have you checked whether it even seems to be true? Try n=m=2.
 
  • #14
haruspex said:
Have you checked whether it even seems to be true? Try n=m=2.
I was about to ask a similar question.

This may also be what tiny-tim was suggesting.Is there a typo in your expression?

In terms of binomial coefficients, what the OP states is that you are trying to prove that

\displaystyle \ \binom{n}{m}=\binom {n} {m-2}+2\binom {n-1} {m-1}+\binom {n-2} {m-2}\ .
 
  • #15
SammyS said:
I was about to ask a similar question.

This may also be what tiny-tim was suggesting.


Is there a typo in your expression?

In terms of binomial coefficients, what the OP states is that you are trying to prove that

\displaystyle \ \binom{n}{m}=\binom {n} {m-2}+2\binom {n-1} {m-1}+\binom {n-2} {m-2}\ .

That is exactly what i need to prove, i just didn't know how to write it like that using Latex. Here is full text of problem:

Prove that \displaystyle \ \binom{n}{m}=\binom {n} {m-2}+2\binom {n-1} {m-1}+\binom {n-2} {m-2}\ n\geq2, m \geq2

When i plug in n=m=2 i get \frac{0}{0} = \frac{2}{0} + \frac{2}{0} + \frac{0}{0}
 
  • #16
Government$ said:
Prove that \displaystyle \ \binom{n}{m}=\binom {n} {m-2}+2\binom {n-1} {m-1}+\binom {n-2} {m-2}\ n\geq2, m \geq2

When i plug in n=m=2 i get \frac{0}{0} = \frac{2}{0} + \frac{2}{0} + \frac{0}{0}
No, 0! = 1. \binom{0}{0} = \binom{2}{0} = 1
 
  • #17
Yea, i overlooked that. With that in mind.

Let n=m=2

\frac{2!}{2!(2-2)!} = \frac{2!}{(2-2)!(2-2 + 2)!} + 2* \frac{(2-1)!}{(2-1)!(2-2)!} + \frac{(2-2)!}{(2-2)!(2-2)!}

\frac{1}{1(0)!} = \frac{1}{(0)!1} + 2* \frac{1}{1(0)!} + \frac{(0)!}{(0)!(0)!}

1 = 1 + 2 + 1

1 = 4
 
  • #18
Government$ said:
That is exactly what i need to prove, i just didn't know how to write it like that using Latex. Here is full text of problem:

Prove that \displaystyle \ \binom{n}{m}=\binom {n} {m-2}+2\binom {n-1} {m-1}+\binom {n-2} {m-2}\ n\geq2, m \geq2

When i plug in n=m=2 i get \frac{0}{0} = \frac{2}{0} + \frac{2}{0} + \frac{0}{0}
Do you suppose the problem should have been to prove

\displaystyle \ \binom{n}{m}=\binom {n-2} {m-2}+2\binom {n-2} {m-1}+\binom {n-2} {m}\,,\ n\geq2, m \geq2 \ ?
 
  • #19
Maybe i will try to slove it,

This problem should be difficult according to my textbook, here is a pic from my textbook:

http://img444.imageshack.us/img444/8316/img00121201304241927.jpg
 
Last edited by a moderator:
  • #20
SammyS said:
Do you suppose the problem should have been to prove

\displaystyle \ \binom{n}{m}=\binom {n-2} {m-2}+2\binom {n-2} {m-1}+\binom {n-2} {m}\,,\ n\geq2, m \geq2 \ ?

I managed to prove this without much problem.
Probably it is a typo in my textbook.
 
Back
Top