How can we use the partition generating function to solve a problem like this?
TO show the number of ways a certain amount of cents can be formed using pennies, nicklels, dimes, and quarters
the hard part about this seems to be the limit of integers (only being able to use 1, 5, 10, and 25)
@DanJS @Kainui @rational @Blacksteel
In my understanding, the general idea is as follows: one writes down the generating polynomial that includes all of the listed amounts of cents as powers: \[g(x)=x+x^5+x^{10}+x^{25}\]
g(x) represents "a coin" Now if someone asks you take k coins and count the number of ways one could get, say, m cents using these k coins, you apply the trick of generating functions. You mutliply k functions g(x): \[g^k(x)=(x+x^5+x^{10}+x^{25})^k\]. You can now multiply this out, perhaps use multinomial fomula to do so. But let us take a concrete example of k=3 coins, and m=25 cents: \[g^3(x)=(x+x^5+x^{10}+x^{25})^3\]
\[g^3(x)=x^{75}+3 x^{60}+3 x^{55}+3 x^{51}+3 x^{45}+6 x^{40}+6 x^{36}+3 x^{35}+6 x^{31}+x^{30}+3 x^{27}+3 x^{25}+3 x^{21}+3 x^{20}+6 x^{16}+x^{15}+3 x^{12}+3 x^{11}+3 x^7+x^3\]
is there for a specific amount of cents though? Like If I asked for the partitions of 26 cents, i don't think there is a restraint on the amount of coins
(hmm, the line shows partially truncated, but maybe others can see it full). So let me say we are looking for m=40 cents (because that is what seems to fit the display). To get the number of ways three coins can be combined into a sum of 40 cents ais now the factor in front of x^{40}, which turns out to be 6. Six ways.
you mean given a desired total amount of cents what is the way to get it using any number of coins? I would think the answer is similar, but you need to add the term x^0 to g(x) to account for "no contribution from a particular coin), like this
\[g(x)=1+x+x^5+x^{10}+x^{25}\]
then you need to choose k at least that of the desired total amount to account for polynomial to attain the sum using minimal denomitation (1 cent), so that would be g^{40}(x) for 40 cents.
Generally you want to find some formula, similar to the multinomial formula, that gives you a prescription for the factor of each term, so you do not really need to multiply everything out. Does this make sense?
im still a little confused.
possible you can demonstrate a full out example for any number of cents?
when you say "any number of cents" are you referring to the total sum (value in cents) or to the number of coins?
i would like it to be the total sum. but if it's a lot more complicated then we can stick with the number of coins
okay we can take an example: 40 cents using any number of coins
ok
so i took the 40th power of the polynomial g(x) = 1+x+x^5+x^10+x^25. The coefficient of the power 40 of that polynomial is astronomical, about 7000 trillion.
which makes sense if one thinks about having 40 1 cent coins permuted all ways, then 35 pennies permuted with 1 nickel, etc.etc.
Normally, the number of coins is given which makes the problem manageable, like I showed further above.
i was hoping for this to be an ordered partition is that possible
the main point is that you get the essence of why this works, then you will solve any similar problem
what precisely do you mean by ordered partition?
if it was ordered wouldnt it be alot more manageable? like you can only arrange 40 pennies once.
thats actually called non-ordered my bad
if we allowed order partitions that formula is generally 2^(x-1)
then generating is usually for non-ordered
actually my comment on permutation was out of place - the arragement of 40 pennies will be counted only once in the total number of ways we can get 40 cents.
right
so let me make sure i understand correctly: for a simpler example how many non-ordered arrangements of pennies and dimes can get you total of 6 cents,
(sorry i mean to say nickel instead of dime) you want to count 6 pennies as one, then 1 nickel + 5 pennies and that's it, i.e. 2 arrangements?
sounds right
obviously dimes and quarters can be used too but they cant in that situation
but if its near impossible. let's just stick with your first idea.
i need to sign off but will keep thinking about this. thanks
no, thank you :) i still have a day to figure this out thanks for taking your time
This seems to have a bunch of useful info and might actually answer your question https://math.berkeley.edu/~mhaiman/math172-spring10/partitions.pdf. I haven't had time to read through but will do so, too.
im reading it and im looking for the part that will specificlaly help me, know where that is?
@satellite73
but im actually getting more and more interested in your first interpretation of the problem as well
would it be (x+x^5+x^10+x^25)^k for any number of cents?
yes, with k coins
oh no wait i think i know what you mean. you put k as the amount of coins and all the exponents are the possible cents u can get with 3 coins, thats why so many exponents are missing because alot of number of cents cannot be made with only 3 coins, correct?
you got it!
im definitely gonna use that for my research paper. however for the final one due in a week, if its possible to solve my interpretation of the problem, that would be nice as well but thanks anyhow
also, do you know how to do that with non-ordered partitions?
Join our real-time social learning platform and learn together with your friends!