A student asked me the other day the following rather homework-looking question: Given a natural number , how many solutions does the equation

have for and natural numbers?

The question has a very easy answer: Simply notice that and that any like this determines a unique such that is a solution. So, there are solutions if is even (as can be any of ), and there are solutions if is odd.

I didn’t tell the student what the answer is, but I asked what he had tried so far. Among what he showed me there was a piece of paper in which somebody else had scribbled

which caught my interest, and is the reason for this posting.

I don’t think the student had seen the connection between this product of two series, let’s call it , and his question. If we denote by the number of solutions to the equation, the series is the generating function of the sequence , i.e.,

To see this, notice that the coefficient of in is precisely the number of ways we can write as a product of a term from the first series and a term from the second one, i.e., it is the number of solutions with and natural numbers to the equation or, equivalently, . That is to say, the coefficient of in is exactly .

This gives us a purely algebraic (analytic?) way of solving the question, even if there is no understanding of how to approach it from a combinatorial point of view.

Both series on the product that makes up are geometric series, so we have

that of course coincides with the formula we obtained earlier by combinatorial considerations.

The question the student had is a very simple example of a problem about integer partitions, a beautiful area of mathematics that I hope I am not misconstruing by thinking of it as a branch of combinatorial number theory. The technique of generating functions is a very useful and powerful combinatorial tool that I have always found quite nice although, granted, its use is a bit of an overkill for the question at hand. At the same time, this technique provides us with a (standard) method for solving any problem of the same kind: For fixed natural numbers , find for each the number of tuples of natural numbers such that

One can then go further to study the much subtler partition function and its relatives.

(And I still don’t know the name of the student, who didn’t bother to introduce himself, and I have no idea who suggested to him to look at to begin with.)

Well, I got the partial solution finally, here is the question that I tried to solve by the equation x + 2y = n, x,y,n > 0

Let n be a positive integer. Harry’s school year has n school days. Harry has budget of exactly $n for buying exactly one snack per day at school. There are only two types of snacks available: M&M for $1. 00 per packet, or a pair of bananas at $2.00 per pair. The following restrictions must apply to Harry.

(1) Harry must spend all $n on snacks during the school year.

(2) Harry does not have to buy a snack each school day.

At the end of the school year, Harry must report how many times he bought M&M, and how many times he bought bananas. How many different reports are possible?

Anyway, thank you for your posting 🙂 It helps me to understand the solution better.

[…] while ago I posted a short note on integer partitions and generating functions. I am adding a link here, as it is obviously connected to the combinatorial applications of power series we have been […]

The technique of almost disjoint forcing was introduced in MR0289291 (44 #6482). Jensen, R. B.; Solovay, R. M. Some applications of almost disjoint sets. In Mathematical Logic and Foundations of Set Theory (Proc. Internat. Colloq., Jerusalem, 1968), pp. 84–104, North-Holland, Amsterdam, 1970. Fix an almost disjoint family $X=(x_\alpha:\alpha

At the moment most of those decisions come from me, at least for computer science papers (those with a 68 class as primary). The practice of having proceedings and final versions of papers is not exclusive to computer science, but this is where it is most common. I've found more often than not that the journal version is significantly different from the […]

The answer is no in general. For instance, by what is essentially an argument of Sierpiński, if $(X,\Sigma,\nu)$ is a $\sigma$-finite continuous measure space, then no non-null subset of $X$ admits a $\nu\times\nu$-measurable well-ordering. The proof is almost verbatim the one here. It is consistent (assuming large cardinals) that there is an extension of Le […]

I assume by $\aleph$ you mean $\mathfrak c$, the cardinality of the continuum. You can build $D$ by transfinite recursion: Well-order the continuum in type $\mathfrak c$. At stage $\alpha$ you add a point of $A_\alpha$ to your set, and one to its complement. You can always do this because at each stage fewer than $\mathfrak c$ many points have been selected. […]

Stefan, "low" cardinalities do not change by passing from $L({\mathbb R})$ to $L({\mathbb R})[{\mathcal U}]$, so the answer to the second question is negative. More precisely: Assume determinacy in $L({\mathbb R})$. Then $2^\omega/E_0$ is a successor cardinal to ${\mathfrak c}$ (This doesn't matter, all we need is that it is strictly larger. T […]

The power of a set is its cardinality. (As opposed to its power set, which is something else.) As you noticed in the comments, Kurepa trees are supposed to have countable levels, although just saying that a tree has size and height $\omega_1$ is not enough to conclude this, so the definition you quoted is incomplete as stated. Usually the convention is that […]

The key problem in the absence of the axiom of replacement is that there may be well-ordered sets $S$ that are too large in the sense that they are longer than any ordinal. In that case, the collection of ordinals isomorphic to an initial segment of $S$ would be the class of all ordinals, which is not a set. For example, with $\omega$ denoting as usual the f […]

R. Solovay proved that the provably $\mathbf\Delta^1_2$ sets are Lebesgue measurable (and have the property of Baire). A set $A$ is provably $\mathbf\Delta^1_2$ iff there is a real $a$, a $\Sigma^1_2$ formula $\phi(x,y)$ and a $\Pi^1_2$ formula $\psi(x,y)$ such that $A=\{t\mid \phi(t,a)\}=\{t\mid\psi(t,a)\}$, and $\mathsf{ZFC}$ proves that $\phi$ and $\psi$ […]

Yes, the suggested rearrangement converges to 0. This is a particular case of a result of Martin Ohm: For $p$ and $q$ positive integers rearrange the sequence $$\left(\frac{(−1)^{n-1}} n\right)_{n\ge 1} $$ by taking the ﬁrst $p$ positive terms, then the ﬁrst $q$ negative terms, then the next $p$ positive terms, then the next $q$ negative terms, and so on. Th […]

Yes, by the incompleteness theorem. An easy argument is to enumerate the sentences in the language of arithmetic. Assign to each node $\sigma $ of the tree $2^{

I’m the student asked the question 😉

Well, I got the partial solution finally, here is the question that I tried to solve by the equation x + 2y = n, x,y,n > 0

Let n be a positive integer. Harry’s school year has n school days. Harry has budget of exactly $n for buying exactly one snack per day at school. There are only two types of snacks available: M&M for $1. 00 per packet, or a pair of bananas at $2.00 per pair. The following restrictions must apply to Harry.

(1) Harry must spend all $n on snacks during the school year.

(2) Harry does not have to buy a snack each school day.

At the end of the school year, Harry must report how many times he bought M&M, and how many times he bought bananas. How many different reports are possible?

Anyway, thank you for your posting 🙂 It helps me to understand the solution better.

Hi Eliot,

I’m glad this helped. Marion mentioned to me the `M&Ms problem’ the other day, I figured this was the same question without distractions.

[…] while ago I posted a short note on integer partitions and generating functions. I am adding a link here, as it is obviously connected to the combinatorial applications of power series we have been […]