Ask your own question, for FREE!
Mathematics 10 Online
OpenStudy (anonymous):

How many compositions of n with k parts are there in which each part is a even number except that a 1 may occur as a part at most once?

OpenStudy (anonymous):

I can't figure out the generating series to solve for this because I don't know how to deal with the 1 occurring at most once part.

OpenStudy (anonymous):

What is a composition in this context?

OpenStudy (anonymous):

A composition of n is (c1,...,ck) such that each ci is a positive integer and c1+...+ck = n

OpenStudy (anonymous):

Interesting.

OpenStudy (anonymous):

so the compositions of 4 are (4),(3,1),(2,2),(1,3),(2,1,1) etc

OpenStudy (anonymous):

So 1 many occur 0 or 1 times.

OpenStudy (anonymous):

yes

OpenStudy (anonymous):

Order of the sum doesn't matter?

OpenStudy (anonymous):

You distinguish (1,3) and (3,1)

OpenStudy (anonymous):

I think those would be distinct compositions

OpenStudy (anonymous):

What reasoning did you use to get the case with no extra conditions?

OpenStudy (anonymous):

well I could use a generating series that looks like this\[\Phi_S(x) = (\sum\limits_{i\geq0}^{}x^{2i+2})^k\]

OpenStudy (anonymous):

That doesn't have \(n\) in it. What does it count?

OpenStudy (anonymous):

then you can simplify that to be \[(x^2(1-x^2))^k\] by using the geometric series then you find the nth coefficient of that. so, \[[x^n](x^2(1-x^2))^k\] where [x^n] is the nth coeffiecient and that gives the answer

OpenStudy (anonymous):

Is there some theorem you are using? Where did you get \(x^{2i+2}\)?

OpenStudy (anonymous):

I think I'm making this a lot more complicated sounding. I'm going to post the example I'm looking at

OpenStudy (anonymous):

https://db.tt/PkWzwRXl https://db.tt/G9FGb8cO

OpenStudy (anonymous):

this is the simpler example for finding the number of k-part compositions of n in which each part is an odd number

OpenStudy (anonymous):

Why did you go with \(2i+2\) instead of just \(2i\), out of curiosity?

OpenStudy (anonymous):

I needed to exclude 0 since it's not a positive integer

OpenStudy (anonymous):

Hmmm, okay we can count the ways in which 1 happens 0 times (all even) Then we can count the ways in which 1 occurs one times. There should be no overlap, right? So add these ones up.

OpenStudy (anonymous):

For 1 to happens once, we simply guarantee it will be in the composition. This means we first subtract it out of \(n\), giving us \(n-1\) and we deal with one less component giving us \(k-1\).

OpenStudy (anonymous):

So find all even \(k-1\) part compositions for \(n-1\), and then stick a \(1\) in it. How many places are there to put in the \(1\)? We we can stick it at the front, or behind any of the other numbers in the composition. Thus we have \(1+(k-21) = k\) places to put the \(1\).

OpenStudy (anonymous):

So assuming \(f(n,k)\) counts even compositions, use \(f(n,k)+kf(n-1,k-1)\)

OpenStudy (anonymous):

Ok, so determined that the number of compositions of n into k parts where each part is even is \[{k+(n/2)-1\choose n/2}\]

OpenStudy (anonymous):

so if \[f(n,k)+kf(n−1,k−1)\] then \[{k+(n/2)-1 \choose n/2}+k{k-1+((n-1)/2)-1 \choose (n-1)/2}\] would be the solution

OpenStudy (anonymous):

You have to worry a bit about whether n is even or not.

OpenStudy (anonymous):

n would be even if all the numbers were even so you couldn't use one 1 in the composition

OpenStudy (anonymous):

That's all the help I can give you. I don't understand anything about these generating series.

OpenStudy (anonymous):

you've been extremely helpful, I've been stuck on this problem for hours and this is the first time it's started making sense. Generating series aren't a requirement of solving these, it's just an algorithmic method. Solving them through logic generally makes them much easier.

OpenStudy (anonymous):

could you explain what you meant when you said that I need to worry a bit about whether n is even? Why is that significant?

OpenStudy (anonymous):

Because putting 3.5 into a binomial can be troublesome

OpenStudy (anonymous):

ok that makes sense I can see where restrictions would be applied to make it work

OpenStudy (anonymous):

Honestly, Thank you so much!

OpenStudy (anonymous):

Wait, if n is even, you can't put only one 1 in it, because it will give you an odd number.

OpenStudy (anonymous):

If n is odd, then you have to use the other method.

OpenStudy (anonymous):

What I mean is, if n is odd, you have to do the n-1, k-1 thing,

OpenStudy (anonymous):

So, rather than adding them together, maybe each one is a separate answer for each case. That would certainly prevent any .5s making their way into the binomial.

OpenStudy (anonymous):

oh that does make a lot of sense. the sum of all even numbers must be even and the sum of all even numbers plus one must be odd

OpenStudy (anonymous):

separating them into cases makes way more sense

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!