Jump to content

Formal power series: Difference between revisions

From Wikipedia, the free encyclopedia
Content deleted Content added
 
(90 intermediate revisions by 42 users not shown)
Line 1: Line 1:
{{Short description|Infinite sum that is considered independently from any notion of convergence}}
{{Refimprove|date=December 2009}}
{{More citations needed|date=December 2009}}
In [[mathematics]], a '''formal power series''' is a generalization of a [[polynomial]], where the number of terms is allowed to be infinite; this implies giving up the possibility of replacing the variable in the polynomial with an arbitrary number. Thus a formal power series differs from a polynomial in that it may have infinitely many terms, and differs from a [[power series]], whose variables can take on numerical values. One way to view a formal power series is as an infinite ordered sequence of numbers. In this case, the powers of the variable are used only to indicate the order of the coefficients, so that the coefficient of <math>x^5</math> is the fifth term in the sequence. In [[combinatorics]], formal power series provide representations of numerical [[sequence]]s and of [[multiset]]s, and for instance allow concise expressions for [[recursion|recursively]] defined sequences regardless of whether the recursion can be explicitly solved; this is known as the method of [[generating function]]s. More generally, formal power series can include series with any finite number of variables, and with coefficients in an arbitrary [[ring (mathematics)|ring]]. Formal power series can be created from [[Taylor polynomial|Taylor polynomials]] using [[formal moduli]].
In [[mathematics]], a '''formal series''' is an infinite sum that is considered independently from any notion of [[convergent series|convergence]], and can be manipulated with the usual algebraic operations on [[series (mathematics)|series]] (addition, subtraction, multiplication, division, [[partial sum]]s, etc.).

A '''formal power series''' is a special kind of formal series, of the form
:<math>\sum_{n=0}^\infty a_nx^n=a_0+a_1x+ a_2x^2+\cdots,</math>
where the <math>a_n,</math> called ''coefficients'', are numbers or, more generally, elements of some [[ring (mathematics)|ring]], and the <math>x^n</math> are formal powers of the symbol <math>x</math> that is called an [[indeterminate (variable)|indeterminate]] or, commonly, a [[variable (mathematics)|variable]]. Hence, power series can be viewed as a generalization of [[polynomial]]s where the number of terms is allowed to be infinite, and differ from usual [[power series]] by the absence of convergence requirements, which implies that a power series may not represent a function of its variable. Formal power series are in [[one to one correspondence]] with their [[sequence (mathematics)|sequence]]s of coefficients, but the two concepts must not be confused, since the operations that can be applied are different.

A formal power series with coefficients in a ring <math>R</math> is called a formal power series over <math>R.</math> The formal power series over a ring <math>R</math> form a ring, commonly denoted by <math>R[[x]].</math> (It can be seen as the [[I-adic topology#Completion|{{math|(''x'')}}-adic completion]] of the [[polynomial ring]] <math>R[x],</math> in the same way as the [[P-adic integer|{{mvar|p}}-adic integers]] are the {{mvar|p}}-adic completion of the ring of the integers.)

Formal powers series in several indeterminates are defined similarly by replacing the powers of a single indeterminate by [[monomial]]s in several indeterminates.

Formal power series are widely used in [[combinatorics]] for representing sequences of integers as [[generating function]]s. In this context, a [[recurrence relation]] between the elements of a sequence may often be interpreted as a [[differential equation]] that the generating function satisfies. This allows using methods of [[complex analysis]] for combinatorial problems (see [[analytic combinatorics]]).


==Introduction==
==Introduction==
Line 7: Line 18:
:<math>A = 1 - 3X + 5X^2 - 7X^3 + 9X^4 - 11X^5 + \cdots.</math>
:<math>A = 1 - 3X + 5X^2 - 7X^3 + 9X^4 - 11X^5 + \cdots.</math>


If we studied this as a power series, its properties would include, for example, that its [[radius of convergence]] is 1. However, as a formal power series, we may ignore this completely; all that is relevant is the sequence of [[coefficient]]s [1, −3, 5, −7, 9, −11, ...]. In other words, a formal power series is an object that just records a sequence of coefficients. It is perfectly acceptable to consider a formal power series with the [[factorial]]s [1, 1, 2, 6, 24, 120, 720, 5040, ... ] as coefficients, even though the corresponding power series diverges for any nonzero value of ''X''.
If we studied this as a power series, its properties would include, for example, that its [[radius of convergence]] is 1 by the [[Cauchy–Hadamard theorem]]. However, as a formal power series, we may ignore this completely; all that is relevant is the sequence of [[coefficient]]s [1, −3, 5, −7, 9, −11, ...]. In other words, a formal power series is an object that just records a sequence of coefficients. It is perfectly acceptable to consider a formal power series with the [[factorial]]s [1, 1, 2, 6, 24, 120, 720, 5040, ... ] as coefficients, even though the corresponding power series diverges for any nonzero value of ''X''.


Arithmetic on formal power series is carried out by simply pretending that the series are polynomials. For example, if
Algebra on formal power series is carried out by simply pretending that the series are polynomials. For example, if


:<math>B = 2X + 4X^3 + 6X^5 + \cdots,</math>
:<math>B = 2X + 4X^3 + 6X^5 + \cdots,</math>
Line 53: Line 64:


====Ring structure====
====Ring structure====
As a set, <math>R\llbracket X\rrbracket</math> can be constructed as the set <math>R^\N</math> of all infinite sequences of elements of <math>R</math>, indexed by the [[natural number]]s (taken to include 0). Designating a sequence whose term at index <math>n</math> is <math>a_n</math> by <math>(a_n)</math>, one defines addition of two such sequences by
As a set, <math>R[[X]]</math> can be constructed as the set <math>R^\N</math> of all infinite sequences of elements of <math>R</math>, indexed by the [[natural number]]s (taken to include 0). Designating a sequence whose term at index <math>n</math> is <math>a_n</math> by <math>(a_n)</math>, one defines addition of two such sequences by


:<math>(a_n)_{n\in\N} + (b_n)_{n\in\N} = \left( a_n + b_n \right)_{n\in\N}</math>
:<math>(a_n)_{n\in\N} + (b_n)_{n\in\N} = \left( a_n + b_n \right)_{n\in\N}</math>
Line 59: Line 70:
and multiplication by
and multiplication by


:<math>(a_n)_{n\in\N} \times (b_n)_{n\in\N} = \left( \sum_{k=0}^n a_k b_{n-k} \right)_{n\in\N}.</math>
:<math>(a_n)_{n\in\N} \times (b_n)_{n\in\N} = \left( \sum_{k=0}^n a_k b_{n-k} \right)_{\!n\in\N}.</math>


This type of product is called the [[Cauchy product]] of the two sequences of coefficients, and is a sort of discrete [[convolution]]. With these operations, <math>R^\N</math> becomes a commutative ring with zero element <math>(0,0,0,\ldots)</math> and multiplicative identity <math>(1,0,0,\ldots)</math>.
This type of product is called the [[Cauchy product]] of the two sequences of coefficients, and is a sort of discrete [[convolution]]. With these operations, <math>R^\N</math> becomes a commutative ring with zero element <math>(0,0,0,\ldots)</math> and multiplicative identity <math>(1,0,0,\ldots)</math>.


The product is in fact the same one used to define the product of polynomials in one indeterminate, which suggests using a similar notation. One embeds <math>R</math> into <math>R\llbracket X\rrbracket</math> by sending any (constant) <math>a \in R</math> to the sequence <math>(a,0,0,\ldots)</math> and designates the sequence <math>(0,1,0,0,\ldots)</math> by <math>X</math>; then using the above definitions every sequence with only finitely many nonzero terms can be expressed in terms of these special elements as
The product is in fact the same one used to define the product of polynomials in one indeterminate, which suggests using a similar notation. One embeds <math>R</math> into <math>R[[X]]</math> by sending any (constant) <math>a \in R</math> to the sequence <math>(a,0,0,\ldots)</math> and designates the sequence <math>(0,1,0,0,\ldots)</math> by <math>X</math>; then using the above definitions every sequence with only finitely many nonzero terms can be expressed in terms of these special elements as


:<math>(a_0, a_1, a_2, \ldots, a_n, 0, 0, \ldots) = a_0 + a_1 X + \cdots + a_n X^n = \sum_{i=0}^n a_i X^i;</math>
:<math>(a_0, a_1, a_2, \ldots, a_n, 0, 0, \ldots) = a_0 + a_1 X + \cdots + a_n X^n = \sum_{i=0}^n a_i X^i;</math>
Line 80: Line 91:
Having stipulated conventionally that
Having stipulated conventionally that


:<math>(a_0, a_1, a_2, a_3, \ldots) = \sum_{i=0}^\infty a_i X^i, \qquad (1)</math>
{{NumBlk|:|<math>(a_0, a_1, a_2, a_3, \ldots) = \sum_{i=0}^\infty a_i X^i,</math>|{{EquationRef|1}}}}


one would like to interpret the right hand side as a well-defined infinite summation. To that end, a notion of convergence in <math>R^\N</math> is defined and a [[topology]] on <math>R^\N</math> is constructed. There are several equivalent ways to define the desired topology.
one would like to interpret the right hand side as a well-defined infinite summation. To that end, a notion of convergence in <math>R^\N</math> is defined and a [[topology]] on <math>R^\N</math> is constructed. There are several equivalent ways to define the desired topology.


* We may give <math>R^\N</math> the [[product topology]], where each copy of <math>R</math> is given the [[discrete topology]].
* We may give <math>R^\N</math> the [[product topology]], where each copy of <math>R</math> is given the [[discrete topology]].

* We may give <math>R^\N</math> the [[I-adic topology]], where <math>I=(X)</math> is the ideal generated by <math>X</math>, which consists of all sequences whose first term <math>a_0</math> is zero.
* We may give <math>R^\N</math> the [[I-adic topology]], where <math>I=(X)</math> is the ideal generated by <math>X</math>, which consists of all sequences whose first term <math>a_0</math> is zero.
* The desired topology could also be derived from the following [[metric space|metric]]. The distance between distinct sequences <math>(a_n), (b_n) \in R^{\N},</math> is defined to be <math display="block">d((a_n), (b_n)) = 2^{-k},</math> where <math>k</math> is the smallest [[natural number]] such that <math>a_k\neq b_k</math>; the distance between two equal sequences is of course zero.


Informally, two sequences <math>(a_n)</math> and <math>(b_n)</math> become closer and closer if and only if more and more of their terms agree exactly. Formally, the sequence of [[partial sum]]s of some infinite summation converges if for every fixed power of <math>X</math> the coefficient stabilizes: there is a point beyond which all further partial sums have the same coefficient. This is clearly the case for the right hand side of ({{EquationNote|1}}), regardless of the values <math>a_n</math>, since inclusion of the term for <math>i=n</math> gives the last (and in fact only) change to the coefficient of <math>X^n</math>. It is also obvious that the [[limit of a sequence|limit]] of the sequence of partial sums is equal to the left hand side.
* The desired topology could also be derived from the following [[metric space|metric]]. The distance between distinct sequences <math>(a_n), (b_n) \in R^{\N},</math> is defined to be
::<math>d((a_n), (b_n)) = 2^{-k},</math>
:where <math>k</math> is the smallest [[natural number]] such that <math>a_k\neq b_k</math>; the distance between two equal sequences is of course zero.

Informally, two sequences <math>\{a_n\}</math> and <math>\{b_n\}</math> become closer and closer if and only if more and more of their terms agree exactly. Formally, the sequence of [[partial sum]]s of some infinite summation converges if for every fixed power of <math>X</math> the coefficient stabilizes: there is a point beyond which all further partial sums have the same coefficient. This is clearly the case for the right hand side of (1), regardless of the values <math>a_n</math>, since inclusion of the term for <math>i=n</math> gives the last (and in fact only) change to the coefficient of <math>X^n</math>. It is also obvious that the [[limit of a sequence|limit]] of the sequence of partial sums is equal to the left hand side.


This topological structure, together with the ring operations described above, form a topological ring. This is called the '''ring of formal power series over <math>R</math>''' and is denoted by <math>R[[X]]</math>. The topology has the useful property that an infinite summation converges if and only if the sequence of its terms converges to 0, which just means that any fixed power of <math>X</math> occurs in only finitely many terms.
This topological structure, together with the ring operations described above, form a topological ring. This is called the '''ring of formal power series over <math>R</math>''' and is denoted by <math>R[[X]]</math>. The topology has the useful property that an infinite summation converges if and only if the sequence of its terms converges to 0, which just means that any fixed power of <math>X</math> occurs in only finitely many terms.
Line 100: Line 107:
:<math>\left(\sum_{i\in\N} a_i X^i\right) \times \left(\sum_{i\in\N} b_i X^i\right) = \sum_{i,j\in\N} a_i b_j X^{i+j},</math>
:<math>\left(\sum_{i\in\N} a_i X^i\right) \times \left(\sum_{i\in\N} b_i X^i\right) = \sum_{i,j\in\N} a_i b_j X^{i+j},</math>


since only finitely many terms on the right affect any fixed <math>X^n</math>. Infinite products are also defined by the topological structure; it can be seen that an infinite product converges if and only if the sequence of its factors converges to 1.
since only finitely many terms on the right affect any fixed <math>X^n</math>. Infinite products are also defined by the topological structure; it can be seen that an infinite product converges if and only if the sequence of its factors converges to 1 (in which case the product is nonzero) or infinitely many factors have no constant term (in which case the product is zero).


==== Alternative topologies ====
==== Alternative topologies ====
The above topology is the [[Comparison of topologies|finest topology]] for which
The above topology is the [[Comparison of topologies|finest topology]] for which


:<math>\sum_{i=0}^\infty a_i X^i</math>
:<math>\sum_{i=0}^\infty a_i X^i</math>


always converges as a summation to the formal power series designated by the same expression, and it often suffices to give a meaning to infinite sums and products, or other kinds of limits that one wishes to use to designate particular formal power series. It can however happen occasionally that one wishes to use a coarser topology, so that certain expressions become convergent that would otherwise diverge. This applies in particular when the base ring <math>R</math> already comes with a topology other than the discrete one, for instance if it is also a ring of formal power series.
always converges as a summation to the formal power series designated by the same expression, and it often suffices to give a meaning to infinite sums and products, or other kinds of limits that one wishes to use to designate particular formal power series. It can however happen occasionally that one wishes to use a coarser topology, so that certain expressions become convergent that would otherwise diverge. This applies in particular when the base ring <math>R</math> already comes with a topology other than the discrete one, for instance if it is also a ring of formal power series.


Consider the ring of formal power series: <math>\Z[[X]][[Y]]</math>; then the topology of above construction only relates to the indeterminate <math>Y</math>, since the topology that was put on <math>\Z[[X]]</math> has been replaced by the discrete topology when defining the topology of the whole ring. So
In the ring of formal power series <math>\Z[[X]][[Y]]</math>, the topology of above construction only relates to the indeterminate <math>Y</math>, since the topology that was put on <math>\Z[[X]]</math> has been replaced by the discrete topology when defining the topology of the whole ring. So


:<math>\sum_{i\in\N}XY^i</math>
:<math>\sum_{i = 0}^\infty XY^i</math>


converges to the power series suggested, which can be written as <math>\tfrac{X}{1-Y}</math>; however the summation
converges (and its sum can be written as <math>\tfrac{X}{1-Y}</math>); however


:<math>\sum_{i\in\N}X^iY</math>
:<math>\sum_{i = 0}^\infty X^i Y</math>


would be considered to be divergent, since every term affects the coefficient of <math>Y</math> (which coefficient is itself a power series in <math>X</math>). This asymmetry disappears if the power series ring in <math>Y</math> is given the product topology where each copy of <math>\Z[[X]]</math> is given its topology as a ring of formal power series rather than the discrete topology. As a consequence, for convergence of a sequence of elements of <math>\Z[[X]][[Y]]</math> it then suffices that the coefficient of each power of <math>Y</math> converges to a formal power series in <math>X</math>, a weaker condition than stabilizing entirely; for instance in the second example given here the coefficient of <math>Y</math>converges to <math>\tfrac{1}{1-X}</math>, so the whole summation converges to <math>\tfrac{Y}{1-X}</math>.
would be considered to be divergent, since every term affects the coefficient of <math>Y</math>. This asymmetry disappears if the power series ring in <math>Y</math> is given the product topology where each copy of <math>\Z[[X]]</math> is given its topology as a ring of formal power series rather than the discrete topology. With this topology, a sequence of elements of <math>\Z[[X]][[Y]]</math> converges if the coefficient of each power of <math>Y</math> converges to a formal power series in <math>X</math>, a weaker condition than stabilizing entirely. For instance, with this topology, in the second example given above, the coefficient of <math>Y</math>converges to <math>\tfrac{1}{1-X}</math>, so the whole summation converges to <math>\tfrac{Y}{1-X}</math>.


This way of defining the topology is in fact the standard one for repeated constructions of rings of formal power series, and gives the same topology as one would get by taking formal power series in all indeterminates at once. In the above example that would mean constructing <math>\Z[[X,Y]],</math> and here a sequence converges if and only if the coefficient of every monomial <math>X^iY^j</math> stabilizes. This topology, which is also the <math>I</math>-adic topology, where <math>I=(X,Y)</math> is the ideal generated by <math>X</math> and <math>Y</math>, still enjoys the property that a summation converges if and only if its terms tend to 0.
This way of defining the topology is in fact the standard one for repeated constructions of rings of formal power series, and gives the same topology as one would get by taking formal power series in all indeterminates at once. In the above example that would mean constructing <math>\Z[[X,Y]]</math> and here a sequence converges if and only if the coefficient of every monomial <math>X^iY^j</math> stabilizes. This topology, which is also the <math>I</math>-adic topology, where <math>I=(X,Y)</math> is the ideal generated by <math>X</math> and <math>Y</math>, still enjoys the property that a summation converges if and only if its terms tend to 0.


The same principle could be used to make other divergent limits converge. For instance in <math>\R[[X]]</math> the limit
The same principle could be used to make other divergent limits converge. For instance in <math>\R[[X]]</math> the limit


:<math>\lim_{n\to\infty}\left(1+\frac{X}{n}\right)^n</math>
:<math>\lim_{n\to\infty}\left(1+\frac{X}{n}\right)^{\!n}</math>


does not exist, so in particular it does not converge to
does not exist, so in particular it does not converge to


:<math>\exp(X)=\sum_{n\in\N}\frac{X^n}{n!}.</math>
:<math>\exp(X) = \sum_{n\in\N}\frac{X^n}{n!}.</math>


This is because for <math>i\geq 2</math> the coefficient <math>\tbinom{n}{i}/n^i</math> of <math>X^i</math> does not stabilize as <math>n\to \infty</math>. It does however converge in the usual topology of <math>\R</math>, and in fact to the coefficient <math>\tfrac{1}{i!}</math> of <math>\exp(X)</math>. Therefore, if one would give <math>\R[[X]]</math> the product topology of <math>\R^\N</math> where the topology of <math>\R</math> is the usual topology rather than the discrete one, then the above limit would converge to <math>\exp(X)</math>. This more permissive approach is not however the standard when considering formal power series, as it would lead to convergence considerations that are as subtle as they are in [[analysis (mathematics)|analysis]], while the philosophy of formal power series is on the contrary to make convergence questions as trivial as they can possibly be. With this topology it would ''not'' be the case that a summation converges if and only if its terms tend to 0.
This is because for <math>i\geq 2</math> the coefficient <math>\tbinom{n}{i}/n^i</math> of <math>X^i</math> does not stabilize as <math>n\to \infty</math>. It does however converge in the usual topology of <math>\R</math>, and in fact to the coefficient <math>\tfrac{1}{i!}</math> of <math>\exp(X)</math>. Therefore, if one would give <math>\R[[X]]</math> the product topology of <math>\R^\N</math> where the topology of <math>\R</math> is the usual topology rather than the discrete one, then the above limit would converge to <math>\exp(X)</math>. This more permissive approach is not however the standard when considering formal power series, as it would lead to convergence considerations that are as subtle as they are in [[analysis (mathematics)|analysis]], while the philosophy of formal power series is on the contrary to make convergence questions as trivial as they can possibly be. With this topology it would ''not'' be the case that a summation converges if and only if its terms tend to 0.
Line 135: Line 142:


* <math>\Phi</math> is an <math>R</math>-algebra homomorphism
* <math>\Phi</math> is an <math>R</math>-algebra homomorphism

* <math>\Phi</math> is continuous
* <math>\Phi</math> is continuous

* <math>\Phi(X)=x</math>.
* <math>\Phi(X)=x</math>.


== Operations on formal power series ==
== Operations on formal power series ==
One can perform algebraic operations on power series to generate new power series.<ref name="Zwillinger_2014">{{cite book |author-first1=Izrail Solomonovich |author-last1=Gradshteyn |author-link1=Izrail Solomonovich Gradshteyn |author-first2=Iosif Moiseevich |author-last2=Ryzhik |author-link2=Iosif Moiseevich Ryzhik |author-first3=Yuri Veniaminovich |author-last3=Geronimus |author-link3 =Yuri Veniaminovich Geronimus |author-first4=Michail Yulyevich |author-last4=Tseytlin |author-link4=Michail Yulyevich Tseytlin |author-first5=Alan |author-last5=Jeffrey |editor1-first=Daniel |editor1-last=Zwillinger |editor2-first=Victor Hugo |editor2-last=Moll |translator=Scripta Technica, Inc. |title=Table of Integrals, Series, and Products |publisher=[[Academic Press, Inc.]] |date=2015 |orig-year=October 2014 |edition=8 |language=English |isbn=978-0-12-384933-5 |lccn=2014010276 <!-- |url=https://books.google.com/books?id=NjnLAwAAQBAJ |access-date=2016-02-21-->|titlelink=Gradshteyn and Ryzhik |chapter=0.313 |page=18}} (Several previous editions as well.)</ref><ref name=formalpowerseries>{{cite journal | first=Ivan | last=Niven | authorlink=Ivan Niven | title=Formal Power Series | journal=[[American Mathematical Monthly]] | volume=76 | issue=8 | date=October 1969 | pages=871–889 | doi=10.1080/00029890.1969.12000359}}</ref> Besides the ring structure operations defined above, we have the following.
One can perform algebraic operations on power series to generate new power series.<ref name="Zwillinger_2014">{{cite book |author-first1=Izrail Solomonovich |author-last1=Gradshteyn |author-link1=Izrail Solomonovich Gradshteyn |author-first2=Iosif Moiseevich |author-last2=Ryzhik |author-link2=Iosif Moiseevich Ryzhik |author-first3=Yuri Veniaminovich |author-last3=Geronimus |author-link3 =Yuri Veniaminovich Geronimus |author-first4=Michail Yulyevich |author-last4=Tseytlin |author-link4=Michail Yulyevich Tseytlin |author-first5=Alan |author-last5=Jeffrey |editor1-first=Daniel |editor1-last=Zwillinger |editor2-first=Victor Hugo |editor2-last=Moll |editor-link2=Victor Hugo Moll |translator=Scripta Technica, Inc. |title=Table of Integrals, Series, and Products |publisher=[[Academic Press, Inc.]] |date=2015 |orig-year=October 2014 |edition=8 |language=en |isbn=978-0-12-384933-5 |lccn=2014010276 <!-- |url=https://books.google.com/books?id=NjnLAwAAQBAJ |access-date=2016-02-21-->|title-link=Gradshteyn and Ryzhik |chapter=0.313 |page=18}} (Several previous editions as well.)</ref><ref name=formalpowerseries>{{cite journal | first=Ivan | last=Niven | author-link=Ivan Niven | title=Formal Power Series | journal=[[American Mathematical Monthly]] | volume=76 | issue=8 | date=October 1969 | pages=871–889 | doi=10.1080/00029890.1969.12000359}}</ref> Besides the ring structure operations defined above, we have the following.


===Power series raised to powers===
===Power series raised to powers===
For any [[natural number]] ''n'' we have
For any [[natural number]] {{math|''n''}}, the {{mvar|n}}th power of a formal power series {{mvar|S}} is defined recursively by
<math display="block">\begin{align}S^1&=S\\

S^n&=S\cdot S^{n-1}\quad\text{for } n>1.\end{align}</math>
:<math> \left( \sum_{k=0}^\infty a_k X^k \right)^n = \sum_{m=0}^\infty c_m X^m,</math>


If {{math|''m''}} and {{math|''a''<sub>0</sub>}} are invertible in the ring of coefficients, one can prove<ref>{{cite arXiv |last=Finkel |first=Hal |title=The differential transformation method and Miller's recurrence |date=2010-07-13 |class=math.CA |eprint=1007.2178}}</ref><ref name="Gould">{{Cite journal |last=Gould |first=H. W. |date=1974 |title=Coefficient Identities for Powers of Taylor and Dirichlet Series |url=https://www.jstor.org/stable/2318904 |journal=The American Mathematical Monthly |volume=81 |issue=1 |pages=3–14 |doi=10.2307/2318904 |jstor=2318904 |issn=0002-9890}}</ref><ref>{{Cite journal |last=Zeilberger |first=Doron |year=1995 |title=The J.C.P. miller recurrence for exponentiating a polynomial, and its q-analog. |url=https://doi.org/10.1080/10236199508808006 |journal=Journal of Difference Equations and Applications |volume=1 |issue=1 |pages=57–60 |doi=10.1080/10236199508808006 |via=Taylor & Francis Online}}</ref>{{Efn|The formula is often attributed to [[J.C.P. Miller]], but it has a long history of rediscovery, dating back to least Euler's discovery in 1748.<ref name="Gould" />}}
<math display="block"> \left( \sum_{k=0}^\infty a_k X^k \right)^{\!n} =\, \sum_{m=0}^\infty c_m X^m,</math>
where
where
<math display="block">\begin{align}

:<math>\begin{align}
c_0 &= a_0^n,\\
c_0 &= a_0^n,\\
c_m &= \frac{1}{m a_0} \sum_{k=1}^m (kn - m+k) a_{k} c_{m-k}, \ \ \ m \geq 1.
c_m &= \frac{1}{m a_0} \sum_{k=1}^m (kn - m+k) a_{k} c_{m-k}, \ \ \ m \geq 1.
\end{align}</math>In the case of formal power series with complex coefficients, the complex powers are well defined for series {{math|''f''}} with constant term equal to {{math|1}}. In this case, <math>f^{\alpha}</math> can be defined either by composition with the [[binomial series]] {{math|(1+''x'')<sup>''α''</sup>}}, or by composition with the exponential and the logarithmic series, <math>f^{\alpha} = \exp(\alpha\log(f)),</math> or as the solution of the differential equation (in terms of series) <math> f(f^{\alpha})' = \alpha f^{\alpha} f'</math> with constant term 1; the three definitions are equivalent. The rules of calculus <math>(f^\alpha)^\beta = f^{\alpha\beta}</math> and <math>f^\alpha g^\alpha = (fg)^\alpha</math> easily follow.
\end{align}</math>

(This formula can only be used if ''m'' and ''a''<sub>0</sub> are invertible in the ring of coefficients.)

In the case of formal power series with complex coefficients, the complex powers are well defined at least for series ''f'' with constant term equal to 1. In this case, <math>f^{\alpha}</math> can be defined either by composition with the [[binomial series]] (1+''x'')<sup>α</sup>, or by composition with the exponential and the logarithmic series, <math>f^{\alpha}=\exp(\alpha\log(f)),</math> or as the solution of the differential equation <math>f( f^{\alpha})' = \alpha f^{\alpha} f'</math> with constant term 1, the three definitions being equivalent. The rules of calculus <math>(f^\alpha)^\beta = f^{\alpha\beta}</math> and <math>f^\alpha g^\alpha = (fg)^\alpha</math> easily follow.


=== Multiplicative inverse ===
=== Multiplicative inverse ===
Line 171: Line 173:
\end{align}</math>
\end{align}</math>


An important special case is that the [[geometric series]] formula is valid in <math>K[[X]]</math>:
An important special case is that the [[geometric series]] formula is valid in <math>R[[X]]</math>:


:<math>(1 - X)^{-1} = \sum_{n=0}^\infty X^n.</math>
:<math>(1 - X)^{-1} = \sum_{n=0}^\infty X^n.</math>
Line 187: Line 189:


=== Extracting coefficients ===
=== Extracting coefficients ===
The coefficient extraction operator applied to a formal power series
The coefficient extraction operator applied to a formal power series


:<math>f(X) = \sum_{n=0}^\infty a_n X^n </math>
:<math>f(X) = \sum_{n=0}^\infty a_n X^n </math>
Line 200: Line 202:


=== Composition ===
=== Composition ===
Given formal power series
Given two formal power series


:<math>f(X) = \sum_{n=1}^\infty a_n X^n = a_1 X + a_2 X^2 + \cdots</math>
:<math>f(X) = \sum_{n=1}^\infty a_n X^n = a_1 X + a_2 X^2 + \cdots</math>
:<math>g(X) = \sum_{n=0}^\infty b_n X^n = b_0 + b_1 X + b_2 X^2 + \cdots,</math>
:<math>g(X) = \sum_{n=0}^\infty b_n X^n = b_0 + b_1 X + b_2 X^2 + \cdots</math>
such that <math>a_0=0,</math>

one may form the ''composition''
one may form the ''composition''


Line 214: Line 216:


Here the sum is extended over all (''k'', ''j'') with <math>k\in\N</math> and <math>j\in\N_+^k</math> with <math>|j|:=j_1+\cdots+j_k=n.</math>
Here the sum is extended over all (''k'', ''j'') with <math>k\in\N</math> and <math>j\in\N_+^k</math> with <math>|j|:=j_1+\cdots+j_k=n.</math>

Since <math>a_0=0,</math> one must have <math>k\le n</math> and <math>j_i\le n</math> for every <math>i. </math> This implies that the above sum is finite and that the coefficient <math>c_n</math> is the coefficient of <math>X^n</math> in the polynomial <math>g_n(f_n(X))</math>, where <math>f_n</math> and <math>g_n</math> are the polynomials obtained by truncating the series at <math>x^n,</math> that is, by removing all terms involving a power of <math>X</math> higher than <math>n.</math>


A more explicit description of these coefficients is provided by [[Faà di Bruno's formula#Formal power series version|Faà di Bruno's formula]], at least in the case where the coefficient ring is a field of [[Characteristic (algebra)|characteristic 0]].
A more explicit description of these coefficients is provided by [[Faà di Bruno's formula#Formal power series version|Faà di Bruno's formula]], at least in the case where the coefficient ring is a field of [[Characteristic (algebra)|characteristic 0]].


A point here is that this operation is only valid when <math>f(X)</math> has ''no constant term'', so that each <math>c_n</math> depends on only a finite number of coefficients of <math>f(X)</math> and <math>g(X)</math>. In other words, the series for <math>g(f(X))</math> converges in the [[Completion (ring theory)#Krull topology|topology]] of <math>R[[X]]</math>.
Composition is only valid when <math>f(X)</math> has ''no constant term'', so that each <math>c_n</math> depends on only a finite number of coefficients of <math>f(X)</math> and <math>g(X)</math>. In other words, the series for <math>g(f(X))</math> converges in the [[Completion (ring theory)#Krull topology|topology]] of <math>R[[X]]</math>.


==== Example ====
==== Example ====
Assume that the ring <math>R</math> has characteristic 0 and the nonzero integers are invertible in <math>R</math>. If we denote by <math>\exp(X)</math> the formal power series
Assume that the ring <math>R</math> has characteristic 0 and the nonzero integers are invertible in <math>R</math>. If one denotes by <math>\exp(X)</math> the formal power series


:<math>\exp(X) = 1 + X + \frac{X^2}{2!} + \frac{X^3}{3!} + \frac{X^4}{4!} + \cdots,</math>
:<math>\exp(X) = 1 + X + \frac{X^2}{2!} + \frac{X^3}{3!} + \frac{X^4}{4!} + \cdots,</math>


then the expression
then the equality


:<math>\exp(\exp(X) - 1) = 1 + X + X^2 + \frac{5X^3}6 + \frac{5X^4}8 + \cdots</math>
:<math>\exp(\exp(X) - 1) = 1 + X + X^2 + \frac{5X^3}6 + \frac{5X^4}8 + \cdots</math>


makes perfect sense as a formal power series. However, the statement
makes perfect sense as a formal power series, since the constant coefficient of <math>\exp(X) - 1</math> is zero.

:<math>\exp(\exp(X)) = e \exp(\exp(X) - 1) = e + eX + eX^2 + \frac{5eX^3}{6} + \cdots</math>

is not a valid application of the composition operation for formal power series. Rather, it is confusing the notions of convergence in <math>R[[X]]</math> and convergence in <math>R</math>; indeed, the ring <math>R</math> may not even contain any number <math>e</math> with the appropriate properties.


=== Composition inverse ===
=== Composition inverse ===
Whenever a formal series
Whenever a formal series


:<math>f(X)=\sum_k f_k X^k \in R[[X]]</math>
:<math>f(X)=\sum_k f_k X^k \in R[[X]]</math>


has ''f''<sub>0</sub> = 0 and ''f''<sub>1</sub> being an invertible element of ''R'', there exists a series
has ''f''<sub>0</sub> = 0 and ''f''<sub>1</sub> being an invertible element of ''R'', there exists a series


:<math>g(X)=\sum_k g_k X^k</math>
:<math>g(X)=\sum_k g_k X^k</math>


that is the [[composition inverse]] of <math>f</math>, meaning that composing <math>f</math> with <math>g</math> gives the series representing the [[identity function]] (whose first coefficient is 1 and all other coefficients are zero). The coefficients of <math>g</math> may be found recursively by using the above formula for the coefficients of a composition, equating them with those of the composition identity ''X'' (that is 1 at degree 1 and 0 at every degree greater than 1). In the case when the coefficient ring is a field of characteristic 0, the [[#The Lagrange inversion formula|Lagrange inversion formula]] provides a powerful tool to compute the coefficients of ''g'', as well as the coefficients of the (multiplicative) powers of ''g''.
that is the [[composition inverse]] of <math>f</math>, meaning that composing <math>f</math> with <math>g</math> gives the series representing the [[identity function]] <math>x = 0 + 1x + 0x^2+ 0x^3+\cdots</math>. The coefficients of <math>g</math> may be found recursively by using the above formula for the coefficients of a composition, equating them with those of the composition identity ''X'' (that is 1 at degree 1 and 0 at every degree greater than 1). In the case when the coefficient ring is a field of characteristic 0, the [[#The Lagrange inversion formula|Lagrange inversion formula]] (discussed below) provides a powerful tool to compute the coefficients of ''g'', as well as the coefficients of the (multiplicative) powers of ''g''.


=== Formal differentiation ===
=== Formal differentiation ===
Line 250: Line 250:
:<math>f = \sum_{n\geq 0} a_n X^n \in R[[X]],</math>
:<math>f = \sum_{n\geq 0} a_n X^n \in R[[X]],</math>


we define its '''[[formal derivative]]''', denoted ''Df'' or ''f''′, by
we define its '''[[formal derivative]]''', denoted ''Df'' or ''f'' ′, by


:<math> Df = \sum_{n \geq 1} a_n n X^{n-1}.</math>
:<math> Df = f' = \sum_{n \geq 1} a_n n X^{n-1}.</math>


The symbol ''D'' is called the '''formal differentiation operator'''. The motivation behind this definition is that it simply mimics term-by-term differentiation of a polynomial.
The symbol ''D'' is called the '''formal differentiation operator'''. This definition simply mimics term-by-term differentiation of a polynomial.


This operation is ''R''-[[linear operator|linear]]:
This operation is ''R''-[[linear operator|linear]]:
Line 262: Line 262:
for any ''a'', ''b'' in ''R'' and any ''f'', ''g'' in <math>R[[X]].</math> Additionally, the formal derivative has many of the properties of the usual [[derivative]] of calculus. For example, the [[product rule]] is valid:
for any ''a'', ''b'' in ''R'' and any ''f'', ''g'' in <math>R[[X]].</math> Additionally, the formal derivative has many of the properties of the usual [[derivative]] of calculus. For example, the [[product rule]] is valid:


:<math>D(fg) = f \cdot (Dg) + (Df) \cdot g,</math>
:<math>D(fg) \ =\ f \cdot (Dg) + (Df) \cdot g,</math>


and the [[chain rule]] works as well:
and the [[chain rule]] works as well:
Line 275: Line 275:


where ''D''<sup>''k''</sup> denotes the ''k''th formal derivative (that is, the result of formally differentiating ''k'' times).
where ''D''<sup>''k''</sup> denotes the ''k''th formal derivative (that is, the result of formally differentiating ''k'' times).

=== Formal antidifferentiation ===
If <math>R</math> is a ring with characteristic zero and the nonzero integers are invertible in <math>R</math>, then given a formal power series

:<math>f = \sum_{n\geq 0} a_n X^n \in R[[X]],</math>

we define its '''formal antiderivative''' or '''formal indefinite integral''' by

:<math> D^{-1} f = \int f\ dX = C + \sum_{n \geq 0} a_n \frac{X^{n+1}}{n+1}.</math>

for any constant <math>C \in R</math>.

This operation is ''R''-[[linear operator|linear]]:

:<math>D^{-1}(af + bg) = a \cdot D^{-1}f + b \cdot D^{-1}g</math>

for any ''a'', ''b'' in ''R'' and any ''f'', ''g'' in <math>R[[X]].</math> Additionally, the formal antiderivative has many of the properties of the usual [[antiderivative]] of calculus. For example, the formal antiderivative is the [[Inverse function#Left and right inverses|right inverse]] of the formal derivative:

:<math>D(D^{-1}(f)) = f</math>

for any <math>f \in R[[X]]</math>.


== Properties ==
== Properties ==
Line 287: Line 308:
Several algebraic properties of <math>R</math> are inherited by <math>R[[X]]</math>:
Several algebraic properties of <math>R</math> are inherited by <math>R[[X]]</math>:


* if <math>R</math> is a [[local ring]], then so is <math>R[[X]]</math>,
* if <math>R</math> is a [[local ring]], then so is <math>R[[X]]</math> (with the set of [[#Multiplicative inverse|non units]] the unique maximal ideal),
* if <math>R</math> is [[noetherian ring|Noetherian]], then so is <math>R[[X]]</math> (a version of the [[Hilbert basis theorem]]),

* if <math>R</math> is [[noetherian ring|Noetherian]], then so is <math>R[[X]]</math>. This is a version of the [[Hilbert basis theorem]],
* if <math>R</math> is an [[integral domain]], then so is <math>R[[X]]</math>, and

* if <math>R</math> is an [[integral domain]], then so is <math>R[[X]]</math>,

* if <math>K</math> is a [[field (mathematics)|field]], then <math>K[[X]]</math> is a [[discrete valuation ring]].
* if <math>K</math> is a [[field (mathematics)|field]], then <math>K[[X]]</math> is a [[discrete valuation ring]].


Line 323: Line 341:


==Interpreting formal power series as functions==
==Interpreting formal power series as functions==
{{Complex analysis sidebar}}
In [[mathematical analysis]], every convergent [[power series]] defines a [[function (mathematics)|function]] with values in the [[real number|real]] or [[complex number|complex]] numbers. Formal power series can also be interpreted as functions, but one has to be careful with the [[function domain|domain]] and [[codomain]]. Let
In [[mathematical analysis]], every convergent [[power series]] defines a [[function (mathematics)|function]] with values in the [[real number|real]] or [[complex number|complex]] numbers. Formal power series over certain special rings can also be interpreted as functions, but one has to be careful with the [[function domain|domain]] and [[codomain]]. Let


:<math>f = \sum a_n X^n \in R[[X]],</math>
:<math>f = \sum a_n X^n \in R[[X]],</math>


and suppose ''S'' is a commutative associative algebra over ''R'', ''I'' is an ideal in ''S'' such that the [[I-adic topology]] on ''S'' is complete, and ''x'' is an element of ''I''. Define:
and suppose <math>S</math> is a commutative associative algebra over <math>R</math>, <math>I</math> is an ideal in <math>S</math> such that the [[I-adic topology]] on <math>S</math> is complete, and <math>x</math> is an element of <math>I</math>. Define:


:<math>f(x) = \sum_{n\ge 0} a_n x^n.</math>
:<math>f(x) = \sum_{n\ge 0} a_n x^n.</math>


This series is guaranteed to converge in ''S'' given the above assumptions on ''x''. Furthermore, we have
This series is guaranteed to converge in <math>S</math> given the above assumptions on <math>x</math>. Furthermore, we have


:<math> (f+g)(x) = f(x) + g(x)</math>
:<math> (f+g)(x) = f(x) + g(x)</math>
Line 341: Line 360:
Unlike in the case of bona fide functions, these formulas are not definitions but have to be proved.
Unlike in the case of bona fide functions, these formulas are not definitions but have to be proved.


Since the topology on <math>R[[X]]</math> is the (''X'')-adic topology and <math>R[[X]]</math> is complete, we can in particular apply power series to other power series, provided that the arguments don't have [[constant coefficients]] (so that they belong to the ideal (''X'')): ''f''(0), ''f''(''X''<sup>2</sup>−''X'') and ''f''((1−''X'')<sup>−1</sup>&nbsp;−&nbsp;1) are all well defined for any formal power series <math>f \in R[[X]].</math>
Since the topology on <math>R[[X]]</math> is the <math>(X)</math>-adic topology and <math>R[[X]]</math> is complete, we can in particular apply power series to other power series, provided that the arguments don't have [[constant coefficients]] (so that they belong to the ideal <math>(X)</math>): <math>f(0)</math>, <math>f(X^2-X)</math> and <math>f((1-X)^{-1}-1)</math> are all well defined for any formal power series <math>f \in R[[X]].</math>


With this formalism, we can give an explicit formula for the multiplicative inverse of a power series ''f'' whose constant coefficient ''a'' = ''f''(0) is invertible in ''R'':
With this formalism, we can give an explicit formula for the multiplicative inverse of a power series <math>f</math> whose constant coefficient <math>a=f(0)</math> is invertible in <math>R</math>:


:<math>f^{-1} = \sum_{n \ge 0} a^{-n-1} (a-f)^n.</math>
:<math>f^{-1} = \sum_{n \ge 0} a^{-n-1} (a-f)^n.</math>


If the formal power series ''g'' with ''g''(0) = 0 is given implicitly by the equation
If the formal power series <math>g</math> with <math>g(0)=0</math> is given implicitly by the equation


:<math>f(g) =X</math>
:<math>f(g) =X</math>


where ''f'' is a known power series with ''f''(0) = 0, then the coefficients of ''g'' can be explicitly computed using the [[#The Lagrange inversion formula|Lagrange inversion formula]].
where <math>f</math> is a known power series with <math>f(0)=0</math>, then the coefficients of <math>g</math> can be explicitly computed using the [[#The Lagrange inversion formula|Lagrange inversion formula]].


==Generalizations==
==Generalizations==


=== Formal Laurent series ===
=== Formal Laurent series ===
A '''formal Laurent series''' over a ring <math>R</math> is defined in a similar way to a formal power series, except that we also allow finitely many terms of negative degree (this is different from the classical [[Laurent series]]), that is series of the form
The '''formal Laurent series''' over a ring <math>R</math> are defined in a similar way to a formal power series, except that we also allow finitely many terms of negative degree. That is, they are the series that can be written as


:<math>f = \sum_{n\in\Z} a_n X^n</math>
:<math>f = \sum_{n = N}^\infty a_n X^n</math>


where '''<math>a_n=0</math> for all but finitely many negative indices <math>n</math>'''. Multiplication of such series can be defined. Indeed, similarly to the definition for formal power series, the coefficient of ''X<sup>k</sup>'' of two series with respective sequences of coefficients <math>\{a_n\}</math> and <math>\{b_n\}</math> is
for some integer <math>N</math>, so that there are only finitely many negative <math>n</math> with <math>a_n \neq 0</math>. (This is different from the classical [[Laurent series]] of [[complex analysis]].) For a non-zero formal Laurent series, the minimal integer <math>n</math> such that <math> a_n\neq 0</math> is called the ''order'' of <math>f</math> and is denoted <math>\operatorname{ord}(f).</math> (The order of the zero series is <math>+\infty</math>.)


Multiplication of such series can be defined. Indeed, similarly to the definition for formal power series, the coefficient of <math>X^k</math> of two series with respective sequences of coefficients <math>\{a_n\}</math> and <math>\{b_n\}</math> is
:<math>\sum_{i\in\Z}a_ib_{k-i},</math>
<math display="block">\sum_{i\in\Z}a_ib_{k-i}.</math>
This sum has only finitely many nonzero terms because of the assumed vanishing of coefficients at sufficiently negative indices.


The formal Laurent series form the '''ring of formal Laurent series''' over <math>R</math>, denoted by <math>R((X))</math>.{{efn|For each nonzero formal Laurent series, the order is an integer (that is, the degrees of the terms are bounded below). But the ring <math>R((X))</math> contains series of all orders.}} It is equal to the [[localization of a ring|localization]] of the ring <math>R[[X]]</math> of formal power series with respect to the set of positive powers of <math>X</math>. If <math>R=K</math> is a [[field (mathematics)|field]], then <math>K((X))</math> is in fact a field, which may alternatively be obtained as the [[field of fractions]] of the [[integral domain]] <math>K[[X]]</math>.
which sum is effectively finite because of the assumed vanishing of coefficients at sufficiently negative indices, and which sum zero for sufficiently negative <math>k</math> for the same reason.


As with <math>R[[X]]</math>, the ring <math>R((X))</math> of formal Laurent series may be endowed with the structure of a topological ring by introducing the metric
For a non-zero formal Laurent series, the minimal integer <math>n</math> such that <math> a_n\neq 0</math> is called the order of <math>f</math>, denoted <math>\operatorname{ord}(f).</math> (The order of the zero series is <math>+\infty</math>.) The formal Laurent series form the '''ring of formal Laurent series''' over <math>R</math>, denoted by <math>R((X))</math>. It is equal to the [[localization of a ring|localization]] of <math>R[[X]]</math> with respect to the set of positive powers of <math>X</math>. It is a topological ring with the metric:
<math display="block">d(f,g)=2^{-\operatorname{ord}(f-g)}.</math>


One may define formal differentiation for formal Laurent series in the natural (term-by-term) way. Precisely, the formal derivative of the formal Laurent series <math>f</math> above is
:<math>d(f,g)=2^{-\operatorname{ord}(f-g)}.</math>
<math display="block">f' = Df = \sum_{n\in\Z} na_n X^{n-1},</math>

which is again a formal Laurent series. If <math>f</math> is a non-constant formal Laurent series and with coefficients in a field of characteristic 0, then one has
If <math>R=K</math> is a [[field (mathematics)|field]], then <math>K((X))</math> is in fact a field, which may alternatively be obtained as the [[field of fractions]] of the [[integral domain]] <math>K[[X]]</math>.
<math display="block">\operatorname{ord}(f')= \operatorname{ord}(f)-1.</math>

However, in general this is not the case since the factor <math>n</math> for the lowest order term could be equal to 0 in <math>R</math>.
One may define formal differentiation for formal Laurent series in a natural way (term-by-term). Precisely, the formal derivative of the formal Laurent series <math>f</math> above is

:<math>f' = Df = \sum_{n\in\Z} na_n X^{n-1}</math>

which is again an element of <math>K((X))</math>. Notice that if <math>f</math> is a non-constant formal Laurent series, and K is a field of characteristic 0, then one has

:<math>\operatorname{ord}(f')= \operatorname{ord}(f)-1.</math>

However, in general this is not the case since the factor ''n'' for the lowest order term could be equal to 0 in ''R''.


====Formal residue====
====Formal residue====
Line 387: Line 401:
:<math>D\colon K((X))\to K((X))</math>
:<math>D\colon K((X))\to K((X))</math>


is a <math>K</math>-[[derivation (abstract algebra)|derivation]] that satisfies
defined above is a <math>K</math>-[[derivation (abstract algebra)|derivation]] that satisfies


:<math>\ker D=K</math>
:<math>\ker D=K</math>
Line 398: Line 412:
is <math>K</math>-linear, and by the above observation one has an [[exact sequence]]
is <math>K</math>-linear, and by the above observation one has an [[exact sequence]]


:<math>0 \to K \to K((X)) \xrightarrow{D} K((X)) \;\xrightarrow{\operatorname{Res}}\; K \to 0.</math>
:<math>0 \to K \to K((X)) \overset{D}{\longrightarrow} K((X)) \;\overset{\operatorname{Res}}{\longrightarrow}\; K \to 0.</math>


'''Some rules of calculus'''. As a quite direct consequence of the above definition, and of the rules of formal derivation, one has, for any <math>f, g\in K((X))</math>
'''Some rules of calculus'''. As a quite direct consequence of the above definition, and of the rules of formal derivation, one has, for any <math>f, g\in K((X))</math>
:i. <math>\operatorname{Res}(f')=0;</math>
<ol style="list-style-type: lower-roman;"><li> <math>\operatorname{Res}(f')=0;</math></li>
:ii. <math>\operatorname{Res}(fg')=-\operatorname{Res}(f'g);</math>
<li> <math>\operatorname{Res}(fg')=-\operatorname{Res}(f'g);</math></li>
:iii. <math>\operatorname{Res}(f'/f)=\operatorname{ord}(f),\qquad \forall f\neq0;</math>
<li> <math>\operatorname{Res}(f'/f)=\operatorname{ord}(f),\qquad \forall f\neq 0;</math></li>
:iv. <math>\operatorname{Res}\left(( f\circ g) g'\right) = \operatorname{ord}(g)\operatorname{Res}(f),</math> if <math>\operatorname{ord}(g)>0;</math>
<li> <math>\operatorname{Res}\left(( g\circ f) f'\right) = \operatorname{ord}(f)\operatorname{Res}(g),</math> if <math>\operatorname{ord}(f)>0;</math></li>
:v. <math>[X^n]f(X)=\operatorname{Res}\left(X^{-n-1}f(X)\right).</math>
<li> <math>[X^n]f(X)=\operatorname{Res}\left(X^{-n-1}f(X)\right).</math></li></ol>


Property (i) is part of the exact sequence above. Property (ii) follows from (i) as applied to <math>(fg)'=f'g+fg'</math>. Property (iii): any <math>f</math> can be written in the form <math>f=X^mg</math>, with <math>m=\operatorname{ord}(f)</math> and <math>\operatorname{ord}(g)=0</math>: then <math>f'/f = mX^{-1}+g'/g.</math> <math>\operatorname{ord}(g)=0</math> implies <math>g</math> is invertible in <math>K[[X]]\subset \operatorname{im}(D) = \ker(\operatorname{Res}),</math> whence <math>\operatorname{Res}(f'/f)=m.</math> Property (iv): Since <math>\operatorname{im}(D) = \ker(\operatorname{Res}),</math> we can write <math>f=f_{-1}X^{-1}+F',</math> with <math>F \in K((X))</math>. Consequently, <math>(f\circ g)g'= f_{-1}g^{-1}g'+(F'\circ g)g' = f_{-1}g'/g + (F \circ g)'</math> and (iv) follows from (i) and (iii). Property (v) is clear from the definition.
Property (i) is part of the exact sequence above. Property (ii) follows from (i) as applied to <math>(fg)'=f'g+fg'</math>. Property (iii): any <math>f</math> can be written in the form <math>f=X^mg</math>, with <math>m=\operatorname{ord}(f)</math> and <math>\operatorname{ord}(g)=0</math>: then <math>f'/f = mX^{-1}+g'/g.</math> <math>\operatorname{ord}(g)=0</math> implies <math>g</math> is invertible in <math>K[[X]]\subset \operatorname{im}(D) = \ker(\operatorname{Res}),</math> whence <math>\operatorname{Res}(f'/f)=m.</math> Property (iv): Since <math>\operatorname{im}(D) = \ker(\operatorname{Res}),</math> we can write <math>g=g_{-1}X^{-1}+G',</math> with <math>G \in K((X))</math>. Consequently, <math>(g\circ f)f'= g_{-1}f^{-1}f'+(G'\circ f)f' = g_{-1}f'/f + (G \circ f)'</math> and (iv) follows from (i) and (iii). Property (v) is clear from the definition.


===The Lagrange inversion formula===
===The Lagrange inversion formula===
{{main|Lagrange inversion theorem}}
{{main|Lagrange inversion theorem}}
As mentioned above, any formal series <math>f \in K[[X]]</math> with ''f''<sub>0</sub> = 0 and ''f''<sub>1</sub> ≠ 0 has a composition inverse <math>g \in K[[X]].</math> The following relation between the coefficients ''g<sup>n</sup>'' and ''f''<sup>−''k''</sup> holds ("{{Visible anchor|Lagrange inversion formula}}"):
As mentioned above, any formal series <math>f \in K[[X]]</math> with ''f''<sub>0</sub> = 0 and ''f''<sub>1</sub> ≠ 0 has a composition inverse <math>g \in K[[X]].</math> The following relation between the coefficients of ''g<sup>n</sup>'' and ''f''<sup>−''k''</sup> holds ("{{Visible anchor|Lagrange inversion formula}}"):


:<math>k[X^k] g^n=n[X^{-n}]f^{-k}.</math>
:<math>k[X^k] g^n=n[X^{-n}]f^{-k}.</math>
Line 419: Line 433:
:<math>[X^k] g=\frac{1}{k} \operatorname{Res}\left( f^{-k}\right).</math>
:<math>[X^k] g=\frac{1}{k} \operatorname{Res}\left( f^{-k}\right).</math>


Since the proof of the Lagrange inversion formula is a very short computation, it is worth reporting one residue-based proof here (a number of different proofs exist,<ref>{{cite book | last1=Richard | first1=Stanley | title=Enumerative combinatorics. Volume 1. | series =Cambridge Stud. Adv. Math. | volume=49 | location=Cambridge | publisher=[[Cambridge University Press]] | year=2012 | isbn=978-1-107-60262-5 | mr=2868112 }}</ref><ref>{{Citation |last1=Ira|first1=Gessel |date=2016 |title=Lagrange inversion |journal=Journal of Combinatorial Theory, Series A |volume=144 |language=en |pages=212–249 |doi=10.1016/j.jcta.2016.06.018 |arxiv=1609.05988|mr=3534068}}</ref><ref>{{Citation |last1=Surya|first1=Erlang |last2=Warnke |first2=Lutz |date=2023 |title=Lagrange Inversion Formula by Induction |journal=The American Mathematical Monthly |volume=130 |issue=10 |language=en |pages=944–948 |doi=10.1080/00029890.2023.2251344 |arxiv=2305.17576|mr=4669236}}</ref> using, e.g., Cauchy's coefficient formula for holomorphic functions, tree-counting arguments, or induction). Noting <math>\operatorname{ord}(f) =1 </math>, we can apply the rules of calculus above, crucially Rule (iv) substituting <math>X \rightsquigarrow f(X)</math>, to get:
Since the proof of the Lagrange inversion formula is a very short computation, it is worth reporting it here. Since <math>\operatorname{ord}(f) =1 </math>, by the above rules of calculus,


:<math>
:<math>
\begin{align}
\begin{align}
k[X^k] g^n & =k\operatorname{Res}\left( g^n X^{-k-1} \right)
k[X^k] g^n &
\ \stackrel{\mathrm{(v)}}=\
k\operatorname{Res}\left( g^n X^{-k-1} \right)
=k\operatorname{Res}\left(X^n f^{-k-1}f\,'\right)
\ \stackrel{\mathrm{(iv)}}=\
k\operatorname{Res}\left(X^n f^{-k-1}f'\right)
=-\operatorname{Res}\left(X^n (f^{-k})^'\right) \\
\ \stackrel{\mathrm{chain}}=\
-\operatorname{Res}\left(X^n (f^{-k})'\right) \\
&
& =\operatorname{Res}\left(\left(X^n\right)' f^{-k}\right)
\ \stackrel{\mathrm{(ii)}}=\
=n\operatorname{Res}\left(X^{n-1}f^{-k}\right)
\operatorname{Res}\left(\left(X^n\right)' f^{-k}\right)
=n[X^{-n}]f^{-k}.
\ \stackrel{\mathrm{chain}}=\
n\operatorname{Res}\left(X^{n-1}f^{-k}\right)
\ \stackrel{\mathrm{(v)}}=\
n[X^{-n}]f^{-k}.
\end{align}
\end{align}
</math>
</math>


'''Generalizations.''' One may observe that the above computation can be repeated plainly in more general settings than ''K''((''X'')): a generalization of the Lagrange inversion formula is already available working in the <math>\Complex((X))</math>-modules <math>X^{\alpha}\Complex((X)),</math> where α is a complex exponent. As a consequence, if ''f'' and ''g'' are as above, with <math>f_1=g_1=1</math>, we can relate the complex powers of ''f''/''X'' and ''g''/''X'': precisely, if α and β are non-zero complex numbers with negative integer sum, <math>m=-\alpha-\beta\in\N,</math> then
'''Generalizations.''' One may observe that the above computation can be repeated plainly in more general settings than ''K''((''X'')): a generalization of the Lagrange inversion formula is already available working in the <math>\Complex((X))</math>-modules <math>X^{\alpha}\Complex((X)),</math> where α is a complex exponent. As a consequence, if ''f'' and ''g'' are as above, with <math>f_1=g_1=1</math>, we can relate the complex powers of ''f'' / ''X'' and ''g'' / ''X'': precisely, if α and β are non-zero complex numbers with negative integer sum, <math>m=-\alpha-\beta\in\N,</math> then


:<math>\frac{1}{\alpha}[X^m]\left( \frac{f}{X} \right)^\alpha=-\frac{1}{\beta}[X^m]\left( \frac{g}{X} \right)^\beta.</math>
:<math>\frac{1}{\alpha}[X^m]\left( \frac{f}{X} \right)^\alpha=-\frac{1}{\beta}[X^m]\left( \frac{g}{X} \right)^\beta.</math>
Line 439: Line 461:


=== Power series in several variables ===
=== Power series in several variables ===
Formal power series in any number of indeterminates (even infinitely many) can be defined. If ''I'' is an index set and ''X<sub>I</sub>'' is the set of indeterminates ''X<sub>i</sub>'' for ''i''∈''I'', then a [[monomial]] ''X''<sup>α</sup> is any finite product of elements of ''X<sub>I</sub>'' (repetitions allowed); a formal power series in ''X<sub>I</sub>'' with coefficients in a ring ''R'' is determined by any mapping from the set of monomials ''X''<sup>α</sup> to a corresponding coefficient ''c''<sub>α</sub>, and is denoted <math>\textstyle\sum_\alpha c_\alpha X^\alpha</math>. The set of all such formal power series is denoted <math>R[[X_I]],</math> and it is given a ring structure by defining
Formal power series in any number of indeterminates (even infinitely many) can be defined. If ''I'' is an index set and ''X<sub>I</sub>'' is the set of indeterminates ''X<sub>i</sub>'' for ''i''∈''I'', then a [[monomial]] ''X''<sup>''α''</sup> is any finite product of elements of ''X<sub>I</sub>'' (repetitions allowed); a formal power series in ''X<sub>I</sub>'' with coefficients in a ring ''R'' is determined by any mapping from the set of monomials ''X''<sup>''α''</sup> to a corresponding coefficient ''c''<sub>''α''</sub>, and is denoted <math display="inline">\sum_\alpha c_\alpha X^\alpha</math>. The set of all such formal power series is denoted <math>R[[X_I]],</math> and it is given a ring structure by defining


:<math>\left(\sum_\alpha c_\alpha X^\alpha\right)+\left(\sum_\alpha d_\alpha X^\alpha \right)= \sum_\alpha (c_\alpha+d_\alpha) X^\alpha</math>
:<math>\left(\sum_\alpha c_\alpha X^\alpha\right)+\left(\sum_\alpha d_\alpha X^\alpha \right)= \sum_\alpha (c_\alpha+d_\alpha) X^\alpha</math>
Line 456: Line 478:


* A series is invertible if and only if its constant term is invertible in ''R''.
* A series is invertible if and only if its constant term is invertible in ''R''.

* The composition ''f''(''g''(''X'')) of two series ''f'' and ''g'' is defined if ''f'' is a series in a single indeterminate, and the constant term of ''g'' is zero. For a series ''f'' in several indeterminates a form of "composition" can similarly be defined, with as many separate series in the place of ''g'' as there are indeterminates.
* The composition ''f''(''g''(''X'')) of two series ''f'' and ''g'' is defined if ''f'' is a series in a single indeterminate, and the constant term of ''g'' is zero. For a series ''f'' in several indeterminates a form of "composition" can similarly be defined, with as many separate series in the place of ''g'' as there are indeterminates.


Line 465: Line 486:


* Φ is an ''R''-algebra homomorphism
* Φ is an ''R''-algebra homomorphism

* Φ is continuous
* Φ is continuous

* Φ(''X''<sub>''i''</sub>) = ''x''<sub>''i''</sub> for ''i'' = 1, ..., ''r''.
* Φ(''X''<sub>''i''</sub>) = ''x''<sub>''i''</sub> for ''i'' = 1, ..., ''r''.


Line 484: Line 503:
{{expand section|sum, product, examples|date=August 2014}}
{{expand section|sum, product, examples|date=August 2014}}


Given an [[Alphabet (formal_languages)|alphabet]] <math>\Sigma</math> and a [[semiring]] <math>S</math>. The formal power series over <math>S</math> supported on the language <math>\Sigma^*</math> is denoted by <math>S\langle\langle \Sigma^*\rangle\rangle</math>. It consists of all mappings <math>r:\Sigma^*\to S</math>, where <math>\Sigma^*</math> is the [[free monoid]] generated by the non-empty set <math>\Sigma</math>.
Given an [[Alphabet (formal languages)|alphabet]] <math>\Sigma</math> and a [[semiring]] <math>S</math>. The formal power series over <math>S</math> supported on the language <math>\Sigma^*</math> is denoted by <math>S\langle\langle \Sigma^*\rangle\rangle</math>. It consists of all mappings <math>r:\Sigma^*\to S</math>, where <math>\Sigma^*</math> is the [[free monoid]] generated by the non-empty set <math>\Sigma</math>.


The elements of <math>S\langle\langle \Sigma^*\rangle\rangle</math> can be written as formal sums
The elements of <math>S\langle\langle \Sigma^*\rangle\rangle</math> can be written as formal sums
Line 511: Line 530:
With these operations <math>(S\langle\langle \Sigma^*\rangle\rangle,+,\cdot,0,\varepsilon)</math> and <math>(S\langle \Sigma^*\rangle, +,\cdot,0,\varepsilon)</math> are semirings, where <math>\varepsilon</math> is the empty word in <math>\Sigma^*</math>.
With these operations <math>(S\langle\langle \Sigma^*\rangle\rangle,+,\cdot,0,\varepsilon)</math> and <math>(S\langle \Sigma^*\rangle, +,\cdot,0,\varepsilon)</math> are semirings, where <math>\varepsilon</math> is the empty word in <math>\Sigma^*</math>.


These formal power series are used to model the behavior of [[weighted automata]], in [[theoretical computer science]], when the coefficients <math>(r,w)</math> of the series are taken to be the weight of a path with label <math>w</math> in the automata.<!-- the correct symbols for the double angled braces are &#10218; and &#10219; but they work poorly in many browsers. Wikipedia's TeX doesn't support \llangle and \rrangle. Also no support for Greek italics in wiki TeX it seems --><ref> Droste, M., & Kuich, W. (2009). Semirings and Formal Power Series. ''Handbook of Weighted Automata'', 3–28. {{doi|10.1007/978-3-642-01492-5_1}}, p. 12</ref>
These formal power series are used to model the behavior of [[weighted automata]], in [[theoretical computer science]], when the coefficients <math>(r,w)</math> of the series are taken to be the weight of a path with label <math>w</math> in the automata.<!-- the correct symbols for the double angled braces are and ; but they work poorly in many browsers. Wikipedia's TeX doesn't support \llangle and \rrangle. Also no support for Greek italics in wiki TeX it seems --><ref>Droste, M., & Kuich, W. (2009). Semirings and Formal Power Series. ''Handbook of Weighted Automata'', 3–28. {{doi|10.1007/978-3-642-01492-5_1}}, p. 12</ref>


===Replacing the index set by an ordered abelian group===
===Replacing the index set by an ordered abelian group===
Line 520: Line 539:
:<math>\sum_{i \in I} a_i X^i </math>
:<math>\sum_{i \in I} a_i X^i </math>


for all such '''I''', with <math>a_i</math> in a commutative ring <math>R</math>, where we assume that for any index set, if all of the <math>a_i</math> are zero then the sum is zero. Then <math>R((G))</math> is the ring of formal power series on <math>G</math>; because of the condition that the indexing set be well-ordered the product is well-defined, and we of course assume that two elements which differ by zero are the same. Sometimes the notation <math>[[R^G]]</math> is used to denote <math>R((G))</math>.<ref>{{cite journal | first1=Khodr | last1=Shamseddine | first2=Martin | last2=Berz | url= http://www.physics.umanitoba.ca/~khodr/Publications/RS-Overview-offprints.pdf | title=Analysis on the Levi-Civita Field: A Brief Overview | journal= Contemporary Mathematics | volume=508 | pages=215–237 | date=2010}}</ref>
for all such '''I''', with <math>a_i</math> in a commutative ring <math>R</math>, where we assume that for any index set, if all of the <math>a_i</math> are zero then the sum is zero. Then <math>R((G))</math> is the ring of formal power series on <math>G</math>; because of the condition that the indexing set be well-ordered the product is well-defined, and we of course assume that two elements which differ by zero are the same. Sometimes the notation <math>[[R^G]]</math> is used to denote <math>R((G))</math>.<ref>{{cite journal | first1=Khodr | last1=Shamseddine | first2=Martin | last2=Berz | url= http://www.physics.umanitoba.ca/~khodr/Publications/RS-Overview-offprints.pdf | title=Analysis on the Levi-Civita Field: A Brief Overview | journal= Contemporary Mathematics | volume=508 | pages=215–237 | date=2010| doi=10.1090/conm/508/10002 | isbn=9780821847404 }}</ref>


Various properties of <math>R</math> transfer to <math>R((G))</math>. If <math>R</math> is a field, then so is <math>R((G))</math>. If <math>R</math> is an ordered field, we can order <math>R((G))</math> by setting any element to have the same sign as its leading coefficient, defined as the least element of the index set '''I''' associated to a non-zero coefficient. Finally if <math>G</math> is a [[divisible group]] and <math>R</math> is a [[real closed field]], then <math>R((G))</math> is a real closed field, and if <math>R</math> is [[algebraically closed]], then so is <math>R((G))</math>.
Various properties of <math>R</math> transfer to <math>R((G))</math>. If <math>R</math> is a field, then so is <math>R((G))</math>. If <math>R</math> is an ordered field, we can order <math>R((G))</math> by setting any element to have the same sign as its leading coefficient, defined as the least element of the index set '''I''' associated to a non-zero coefficient. Finally if <math>G</math> is a [[divisible group]] and <math>R</math> is a [[real closed field]], then <math>R((G))</math> is a real closed field, and if <math>R</math> is [[algebraically closed]], then so is <math>R((G))</math>.
Line 531: Line 550:
* [[Puiseux series]] are an extension of formal Laurent series, allowing fractional exponents
* [[Puiseux series]] are an extension of formal Laurent series, allowing fractional exponents
* [[Rational series]]
* [[Rational series]]

== See also ==
* [[Ring of restricted power series]]


==Notes==
==Notes==
{{Reflist}}
{{notelist}}


== References ==
== References ==
{{Reflist}}
* {{cite book | last1=Berstel | first1=Jean | author-link1=Jean Berstel | last2=Reutenauer | first2=Christophe | title=Noncommutative rational series with applications | series =Encyclopedia of Mathematics and Its Applications | volume=137 | location=Cambridge | publisher=[[Cambridge University Press]] | year=2011 | isbn=978-0-521-19022-0 | zbl=1250.68007 }}
* {{cite book | last1=Berstel | first1=Jean | author-link1=Jean Berstel | last2=Reutenauer | first2=Christophe | title=Noncommutative rational series with applications | series =Encyclopedia of Mathematics and Its Applications | volume=137 | location=Cambridge | publisher=[[Cambridge University Press]] | year=2011 | isbn=978-0-521-19022-0 | zbl=1250.68007 }}
* [[Nicolas Bourbaki]]: ''Algebra'', IV, §4. Springer-Verlag 1988.
* [[Nicolas Bourbaki]]: ''Algebra'', IV, §4. Springer-Verlag 1988.


==Further reading==
== Further reading ==
* W. Kuich. Semirings and formal power series: Their relevance to formal languages and automata theory. In G. Rozenberg and A. Salomaa, editors, Handbook of Formal Languages, volume 1, Chapter 9, pages 609–677. Springer, Berlin, 1997, {{ISBN|3-540-60420-0}}
* W. Kuich. Semirings and formal power series: Their relevance to formal languages and automata theory. In G. Rozenberg and A. Salomaa, editors, Handbook of Formal Languages, volume 1, Chapter 9, pages 609–677. Springer, Berlin, 1997, {{ISBN|3-540-60420-0}}
* Droste, M., & Kuich, W. (2009). Semirings and Formal Power Series. ''Handbook of Weighted Automata'', 3–28. {{doi|10.1007/978-3-642-01492-5_1}}
* Droste, M., & Kuich, W. (2009). Semirings and Formal Power Series. ''Handbook of Weighted Automata'', 3–28. {{doi|10.1007/978-3-642-01492-5_1}}
* {{cite book | author=Arto Salomaa | author-link=Arto Salomaa |contribution=Formal Languages and Power Series | pages=103&ndash;132 | isbn=0-444-88074-7 | editor=Jan van Leeuwen | editor-link=Jan van Leeuwen | title=Formal Models and Semantics | publisher=Elsevier | series=Handbook of Theoretical Computer Science | volume=B | year=1990}}


{{series (mathematics)}}
{{series (mathematics)}}



Latest revision as of 06:01, 7 November 2024

In mathematics, a formal series is an infinite sum that is considered independently from any notion of convergence, and can be manipulated with the usual algebraic operations on series (addition, subtraction, multiplication, division, partial sums, etc.).

A formal power series is a special kind of formal series, of the form

where the called coefficients, are numbers or, more generally, elements of some ring, and the are formal powers of the symbol that is called an indeterminate or, commonly, a variable. Hence, power series can be viewed as a generalization of polynomials where the number of terms is allowed to be infinite, and differ from usual power series by the absence of convergence requirements, which implies that a power series may not represent a function of its variable. Formal power series are in one to one correspondence with their sequences of coefficients, but the two concepts must not be confused, since the operations that can be applied are different.

A formal power series with coefficients in a ring is called a formal power series over The formal power series over a ring form a ring, commonly denoted by (It can be seen as the (x)-adic completion of the polynomial ring in the same way as the p-adic integers are the p-adic completion of the ring of the integers.)

Formal powers series in several indeterminates are defined similarly by replacing the powers of a single indeterminate by monomials in several indeterminates.

Formal power series are widely used in combinatorics for representing sequences of integers as generating functions. In this context, a recurrence relation between the elements of a sequence may often be interpreted as a differential equation that the generating function satisfies. This allows using methods of complex analysis for combinatorial problems (see analytic combinatorics).

Introduction

[edit]

A formal power series can be loosely thought of as an object that is like a polynomial, but with infinitely many terms. Alternatively, for those familiar with power series (or Taylor series), one may think of a formal power series as a power series in which we ignore questions of convergence by not assuming that the variable X denotes any numerical value (not even an unknown value). For example, consider the series

If we studied this as a power series, its properties would include, for example, that its radius of convergence is 1 by the Cauchy–Hadamard theorem. However, as a formal power series, we may ignore this completely; all that is relevant is the sequence of coefficients [1, −3, 5, −7, 9, −11, ...]. In other words, a formal power series is an object that just records a sequence of coefficients. It is perfectly acceptable to consider a formal power series with the factorials [1, 1, 2, 6, 24, 120, 720, 5040, ... ] as coefficients, even though the corresponding power series diverges for any nonzero value of X.

Algebra on formal power series is carried out by simply pretending that the series are polynomials. For example, if

then we add A and B term by term:

We can multiply formal power series, again just by treating them as polynomials (see in particular Cauchy product):

Notice that each coefficient in the product AB only depends on a finite number of coefficients of A and B. For example, the X5 term is given by

For this reason, one may multiply formal power series without worrying about the usual questions of absolute, conditional and uniform convergence which arise in dealing with power series in the setting of analysis.

Once we have defined multiplication for formal power series, we can define multiplicative inverses as follows. The multiplicative inverse of a formal power series A is a formal power series C such that AC = 1, provided that such a formal power series exists. It turns out that if A has a multiplicative inverse, it is unique, and we denote it by A−1. Now we can define division of formal power series by defining B/A to be the product BA−1, provided that the inverse of A exists. For example, one can use the definition of multiplication above to verify the familiar formula

An important operation on formal power series is coefficient extraction. In its most basic form, the coefficient extraction operator applied to a formal power series in one variable extracts the coefficient of the th power of the variable, so that and . Other examples include

Similarly, many other operations that are carried out on polynomials can be extended to the formal power series setting, as explained below.

The ring of formal power series

[edit]

If one considers the set of all formal power series in X with coefficients in a commutative ring R, the elements of this set collectively constitute another ring which is written and called the ring of formal power series in the variable X over R.

Definition of the formal power series ring

[edit]

One can characterize abstractly as the completion of the polynomial ring equipped with a particular metric. This automatically gives the structure of a topological ring (and even of a complete metric space). But the general construction of a completion of a metric space is more involved than what is needed here, and would make formal power series seem more complicated than they are. It is possible to describe more explicitly, and define the ring structure and topological structure separately, as follows.

Ring structure

[edit]

As a set, can be constructed as the set of all infinite sequences of elements of , indexed by the natural numbers (taken to include 0). Designating a sequence whose term at index is by , one defines addition of two such sequences by

and multiplication by

This type of product is called the Cauchy product of the two sequences of coefficients, and is a sort of discrete convolution. With these operations, becomes a commutative ring with zero element and multiplicative identity .

The product is in fact the same one used to define the product of polynomials in one indeterminate, which suggests using a similar notation. One embeds into by sending any (constant) to the sequence and designates the sequence by ; then using the above definitions every sequence with only finitely many nonzero terms can be expressed in terms of these special elements as

these are precisely the polynomials in . Given this, it is quite natural and convenient to designate a general sequence by the formal expression , even though the latter is not an expression formed by the operations of addition and multiplication defined above (from which only finite sums can be constructed). This notational convention allows reformulation of the above definitions as

and

which is quite convenient, but one must be aware of the distinction between formal summation (a mere convention) and actual addition.

Topological structure

[edit]

Having stipulated conventionally that

(1)

one would like to interpret the right hand side as a well-defined infinite summation. To that end, a notion of convergence in is defined and a topology on is constructed. There are several equivalent ways to define the desired topology.

  • We may give the product topology, where each copy of is given the discrete topology.
  • We may give the I-adic topology, where is the ideal generated by , which consists of all sequences whose first term is zero.
  • The desired topology could also be derived from the following metric. The distance between distinct sequences is defined to be where is the smallest natural number such that ; the distance between two equal sequences is of course zero.

Informally, two sequences and become closer and closer if and only if more and more of their terms agree exactly. Formally, the sequence of partial sums of some infinite summation converges if for every fixed power of the coefficient stabilizes: there is a point beyond which all further partial sums have the same coefficient. This is clearly the case for the right hand side of (1), regardless of the values , since inclusion of the term for gives the last (and in fact only) change to the coefficient of . It is also obvious that the limit of the sequence of partial sums is equal to the left hand side.

This topological structure, together with the ring operations described above, form a topological ring. This is called the ring of formal power series over and is denoted by . The topology has the useful property that an infinite summation converges if and only if the sequence of its terms converges to 0, which just means that any fixed power of occurs in only finitely many terms.

The topological structure allows much more flexible usage of infinite summations. For instance the rule for multiplication can be restated simply as

since only finitely many terms on the right affect any fixed . Infinite products are also defined by the topological structure; it can be seen that an infinite product converges if and only if the sequence of its factors converges to 1 (in which case the product is nonzero) or infinitely many factors have no constant term (in which case the product is zero).

Alternative topologies

[edit]

The above topology is the finest topology for which

always converges as a summation to the formal power series designated by the same expression, and it often suffices to give a meaning to infinite sums and products, or other kinds of limits that one wishes to use to designate particular formal power series. It can however happen occasionally that one wishes to use a coarser topology, so that certain expressions become convergent that would otherwise diverge. This applies in particular when the base ring already comes with a topology other than the discrete one, for instance if it is also a ring of formal power series.

In the ring of formal power series , the topology of above construction only relates to the indeterminate , since the topology that was put on has been replaced by the discrete topology when defining the topology of the whole ring. So

converges (and its sum can be written as ); however

would be considered to be divergent, since every term affects the coefficient of . This asymmetry disappears if the power series ring in is given the product topology where each copy of is given its topology as a ring of formal power series rather than the discrete topology. With this topology, a sequence of elements of converges if the coefficient of each power of converges to a formal power series in , a weaker condition than stabilizing entirely. For instance, with this topology, in the second example given above, the coefficient of converges to , so the whole summation converges to .

This way of defining the topology is in fact the standard one for repeated constructions of rings of formal power series, and gives the same topology as one would get by taking formal power series in all indeterminates at once. In the above example that would mean constructing and here a sequence converges if and only if the coefficient of every monomial stabilizes. This topology, which is also the -adic topology, where is the ideal generated by and , still enjoys the property that a summation converges if and only if its terms tend to 0.

The same principle could be used to make other divergent limits converge. For instance in the limit

does not exist, so in particular it does not converge to

This is because for the coefficient of does not stabilize as . It does however converge in the usual topology of , and in fact to the coefficient of . Therefore, if one would give the product topology of where the topology of is the usual topology rather than the discrete one, then the above limit would converge to . This more permissive approach is not however the standard when considering formal power series, as it would lead to convergence considerations that are as subtle as they are in analysis, while the philosophy of formal power series is on the contrary to make convergence questions as trivial as they can possibly be. With this topology it would not be the case that a summation converges if and only if its terms tend to 0.

Universal property

[edit]

The ring may be characterized by the following universal property. If is a commutative associative algebra over , if is an ideal of such that the -adic topology on is complete, and if is an element of , then there is a unique with the following properties:

  • is an -algebra homomorphism
  • is continuous
  • .

Operations on formal power series

[edit]

One can perform algebraic operations on power series to generate new power series.[1][2] Besides the ring structure operations defined above, we have the following.

Power series raised to powers

[edit]

For any natural number n, the nth power of a formal power series S is defined recursively by

If m and a0 are invertible in the ring of coefficients, one can prove[3][4][5][a] where In the case of formal power series with complex coefficients, the complex powers are well defined for series f with constant term equal to 1. In this case, can be defined either by composition with the binomial series (1+x)α, or by composition with the exponential and the logarithmic series, or as the solution of the differential equation (in terms of series) with constant term 1; the three definitions are equivalent. The rules of calculus and easily follow.

Multiplicative inverse

[edit]

The series

is invertible in if and only if its constant coefficient is invertible in . This condition is necessary, for the following reason: if we suppose that has an inverse then the constant term of is the constant term of the identity series, i.e. it is 1. This condition is also sufficient; we may compute the coefficients of the inverse series via the explicit recursive formula

An important special case is that the geometric series formula is valid in :

If is a field, then a series is invertible if and only if the constant term is non-zero, i.e. if and only if the series is not divisible by . This means that is a discrete valuation ring with uniformizing parameter .

Division

[edit]

The computation of a quotient

assuming the denominator is invertible (that is, is invertible in the ring of scalars), can be performed as a product and the inverse of , or directly equating the coefficients in :

Extracting coefficients

[edit]

The coefficient extraction operator applied to a formal power series

in X is written

and extracts the coefficient of Xm, so that

Composition

[edit]

Given two formal power series

such that one may form the composition

where the coefficients cn are determined by "expanding out" the powers of f(X):

Here the sum is extended over all (k, j) with and with

Since one must have and for every This implies that the above sum is finite and that the coefficient is the coefficient of in the polynomial , where and are the polynomials obtained by truncating the series at that is, by removing all terms involving a power of higher than

A more explicit description of these coefficients is provided by Faà di Bruno's formula, at least in the case where the coefficient ring is a field of characteristic 0.

Composition is only valid when has no constant term, so that each depends on only a finite number of coefficients of and . In other words, the series for converges in the topology of .

Example

[edit]

Assume that the ring has characteristic 0 and the nonzero integers are invertible in . If one denotes by the formal power series

then the equality

makes perfect sense as a formal power series, since the constant coefficient of is zero.

Composition inverse

[edit]

Whenever a formal series

has f0 = 0 and f1 being an invertible element of R, there exists a series

that is the composition inverse of , meaning that composing with gives the series representing the identity function . The coefficients of may be found recursively by using the above formula for the coefficients of a composition, equating them with those of the composition identity X (that is 1 at degree 1 and 0 at every degree greater than 1). In the case when the coefficient ring is a field of characteristic 0, the Lagrange inversion formula (discussed below) provides a powerful tool to compute the coefficients of g, as well as the coefficients of the (multiplicative) powers of g.

Formal differentiation

[edit]

Given a formal power series

we define its formal derivative, denoted Df or f ′, by

The symbol D is called the formal differentiation operator. This definition simply mimics term-by-term differentiation of a polynomial.

This operation is R-linear:

for any a, b in R and any f, g in Additionally, the formal derivative has many of the properties of the usual derivative of calculus. For example, the product rule is valid:

and the chain rule works as well:

whenever the appropriate compositions of series are defined (see above under composition of series).

Thus, in these respects formal power series behave like Taylor series. Indeed, for the f defined above, we find that

where Dk denotes the kth formal derivative (that is, the result of formally differentiating k times).

Formal antidifferentiation

[edit]

If is a ring with characteristic zero and the nonzero integers are invertible in , then given a formal power series

we define its formal antiderivative or formal indefinite integral by

for any constant .

This operation is R-linear:

for any a, b in R and any f, g in Additionally, the formal antiderivative has many of the properties of the usual antiderivative of calculus. For example, the formal antiderivative is the right inverse of the formal derivative:

for any .

Properties

[edit]

Algebraic properties of the formal power series ring

[edit]

is an associative algebra over which contains the ring of polynomials over ; the polynomials correspond to the sequences which end in zeros.

The Jacobson radical of is the ideal generated by and the Jacobson radical of ; this is implied by the element invertibility criterion discussed above.

The maximal ideals of all arise from those in in the following manner: an ideal of is maximal if and only if is a maximal ideal of and is generated as an ideal by and .

Several algebraic properties of are inherited by :

  • if is a local ring, then so is (with the set of non units the unique maximal ideal),
  • if is Noetherian, then so is (a version of the Hilbert basis theorem),
  • if is an integral domain, then so is , and
  • if is a field, then is a discrete valuation ring.

Topological properties of the formal power series ring

[edit]

The metric space is complete.

The ring is compact if and only if R is finite. This follows from Tychonoff's theorem and the characterisation of the topology on as a product topology.

Weierstrass preparation

[edit]

The ring of formal power series with coefficients in a complete local ring satisfies the Weierstrass preparation theorem.

Applications

[edit]

Formal power series can be used to solve recurrences occurring in number theory and combinatorics. For an example involving finding a closed form expression for the Fibonacci numbers, see the article on Examples of generating functions.

One can use formal power series to prove several relations familiar from analysis in a purely algebraic setting. Consider for instance the following elements of :

Then one can show that

The last one being valid in the ring

For K a field, the ring is often used as the "standard, most general" complete local ring over K in algebra.

Interpreting formal power series as functions

[edit]

In mathematical analysis, every convergent power series defines a function with values in the real or complex numbers. Formal power series over certain special rings can also be interpreted as functions, but one has to be careful with the domain and codomain. Let

and suppose is a commutative associative algebra over , is an ideal in such that the I-adic topology on is complete, and is an element of . Define:

This series is guaranteed to converge in given the above assumptions on . Furthermore, we have

and

Unlike in the case of bona fide functions, these formulas are not definitions but have to be proved.

Since the topology on is the -adic topology and is complete, we can in particular apply power series to other power series, provided that the arguments don't have constant coefficients (so that they belong to the ideal ): , and are all well defined for any formal power series

With this formalism, we can give an explicit formula for the multiplicative inverse of a power series whose constant coefficient is invertible in :

If the formal power series with is given implicitly by the equation

where is a known power series with , then the coefficients of can be explicitly computed using the Lagrange inversion formula.

Generalizations

[edit]

Formal Laurent series

[edit]

The formal Laurent series over a ring are defined in a similar way to a formal power series, except that we also allow finitely many terms of negative degree. That is, they are the series that can be written as

for some integer , so that there are only finitely many negative with . (This is different from the classical Laurent series of complex analysis.) For a non-zero formal Laurent series, the minimal integer such that is called the order of and is denoted (The order of the zero series is .)

Multiplication of such series can be defined. Indeed, similarly to the definition for formal power series, the coefficient of of two series with respective sequences of coefficients and is This sum has only finitely many nonzero terms because of the assumed vanishing of coefficients at sufficiently negative indices.

The formal Laurent series form the ring of formal Laurent series over , denoted by .[b] It is equal to the localization of the ring of formal power series with respect to the set of positive powers of . If is a field, then is in fact a field, which may alternatively be obtained as the field of fractions of the integral domain .

As with , the ring of formal Laurent series may be endowed with the structure of a topological ring by introducing the metric

One may define formal differentiation for formal Laurent series in the natural (term-by-term) way. Precisely, the formal derivative of the formal Laurent series above is which is again a formal Laurent series. If is a non-constant formal Laurent series and with coefficients in a field of characteristic 0, then one has However, in general this is not the case since the factor for the lowest order term could be equal to 0 in .

Formal residue

[edit]

Assume that is a field of characteristic 0. Then the map

defined above is a -derivation that satisfies

The latter shows that the coefficient of in is of particular interest; it is called formal residue of and denoted . The map

is -linear, and by the above observation one has an exact sequence

Some rules of calculus. As a quite direct consequence of the above definition, and of the rules of formal derivation, one has, for any

  1. if

Property (i) is part of the exact sequence above. Property (ii) follows from (i) as applied to . Property (iii): any can be written in the form , with and : then implies is invertible in whence Property (iv): Since we can write with . Consequently, and (iv) follows from (i) and (iii). Property (v) is clear from the definition.

The Lagrange inversion formula

[edit]

As mentioned above, any formal series with f0 = 0 and f1 ≠ 0 has a composition inverse The following relation between the coefficients of gn and fk holds ("Lagrange inversion formula"):

In particular, for n = 1 and all k ≥ 1,

Since the proof of the Lagrange inversion formula is a very short computation, it is worth reporting one residue-based proof here (a number of different proofs exist,[6][7][8] using, e.g., Cauchy's coefficient formula for holomorphic functions, tree-counting arguments, or induction). Noting , we can apply the rules of calculus above, crucially Rule (iv) substituting , to get:

Generalizations. One may observe that the above computation can be repeated plainly in more general settings than K((X)): a generalization of the Lagrange inversion formula is already available working in the -modules where α is a complex exponent. As a consequence, if f and g are as above, with , we can relate the complex powers of f / X and g / X: precisely, if α and β are non-zero complex numbers with negative integer sum, then

For instance, this way one finds the power series for complex powers of the Lambert function.

Power series in several variables

[edit]

Formal power series in any number of indeterminates (even infinitely many) can be defined. If I is an index set and XI is the set of indeterminates Xi for iI, then a monomial Xα is any finite product of elements of XI (repetitions allowed); a formal power series in XI with coefficients in a ring R is determined by any mapping from the set of monomials Xα to a corresponding coefficient cα, and is denoted . The set of all such formal power series is denoted and it is given a ring structure by defining

and

Topology

[edit]

The topology on is such that a sequence of its elements converges only if for each monomial Xα the corresponding coefficient stabilizes. If I is finite, then this the J-adic topology, where J is the ideal of generated by all the indeterminates in XI. This does not hold if I is infinite. For example, if then the sequence with does not converge with respect to any J-adic topology on R, but clearly for each monomial the corresponding coefficient stabilizes.

As remarked above, the topology on a repeated formal power series ring like is usually chosen in such a way that it becomes isomorphic as a topological ring to

Operations

[edit]

All of the operations defined for series in one variable may be extended to the several variables case.

  • A series is invertible if and only if its constant term is invertible in R.
  • The composition f(g(X)) of two series f and g is defined if f is a series in a single indeterminate, and the constant term of g is zero. For a series f in several indeterminates a form of "composition" can similarly be defined, with as many separate series in the place of g as there are indeterminates.

In the case of the formal derivative, there are now separate partial derivative operators, which differentiate with respect to each of the indeterminates. They all commute with each other.

Universal property

[edit]

In the several variables case, the universal property characterizing becomes the following. If S is a commutative associative algebra over R, if I is an ideal of S such that the I-adic topology on S is complete, and if x1, ..., xr are elements of I, then there is a unique map with the following properties:

  • Φ is an R-algebra homomorphism
  • Φ is continuous
  • Φ(Xi) = xi for i = 1, ..., r.

Non-commuting variables

[edit]

The several variable case can be further generalised by taking non-commuting variables Xi for iI, where I is an index set and then a monomial Xα is any word in the XI; a formal power series in XI with coefficients in a ring R is determined by any mapping from the set of monomials Xα to a corresponding coefficient cα, and is denoted . The set of all such formal power series is denoted R«XI», and it is given a ring structure by defining addition pointwise

and multiplication by

where · denotes concatenation of words. These formal power series over R form the Magnus ring over R.[9][10]

On a semiring

[edit]

Given an alphabet and a semiring . The formal power series over supported on the language is denoted by . It consists of all mappings , where is the free monoid generated by the non-empty set .

The elements of can be written as formal sums

where denotes the value of at the word . The elements are called the coefficients of .

For the support of is the set

A series where every coefficient is either or is called the characteristic series of its support.

The subset of consisting of all series with a finite support is denoted by and called polynomials.

For and , the sum is defined by

The (Cauchy) product is defined by

The Hadamard product is defined by

And the products by a scalar and by

and , respectively.

With these operations and are semirings, where is the empty word in .

These formal power series are used to model the behavior of weighted automata, in theoretical computer science, when the coefficients of the series are taken to be the weight of a path with label in the automata.[11]

Replacing the index set by an ordered abelian group

[edit]

Suppose is an ordered abelian group, meaning an abelian group with a total ordering respecting the group's addition, so that if and only if for all . Let I be a well-ordered subset of , meaning I contains no infinite descending chain. Consider the set consisting of

for all such I, with in a commutative ring , where we assume that for any index set, if all of the are zero then the sum is zero. Then is the ring of formal power series on ; because of the condition that the indexing set be well-ordered the product is well-defined, and we of course assume that two elements which differ by zero are the same. Sometimes the notation is used to denote .[12]

Various properties of transfer to . If is a field, then so is . If is an ordered field, we can order by setting any element to have the same sign as its leading coefficient, defined as the least element of the index set I associated to a non-zero coefficient. Finally if is a divisible group and is a real closed field, then is a real closed field, and if is algebraically closed, then so is .

This theory is due to Hans Hahn, who also showed that one obtains subfields when the number of (non-zero) terms is bounded by some fixed infinite cardinality.

[edit]

See also

[edit]

Notes

[edit]
  1. ^ The formula is often attributed to J.C.P. Miller, but it has a long history of rediscovery, dating back to least Euler's discovery in 1748.[4]
  2. ^ For each nonzero formal Laurent series, the order is an integer (that is, the degrees of the terms are bounded below). But the ring contains series of all orders.

References

[edit]
  1. ^ Gradshteyn, Izrail Solomonovich; Ryzhik, Iosif Moiseevich; Geronimus, Yuri Veniaminovich; Tseytlin, Michail Yulyevich; Jeffrey, Alan (2015) [October 2014]. "0.313". In Zwillinger, Daniel; Moll, Victor Hugo (eds.). Table of Integrals, Series, and Products. Translated by Scripta Technica, Inc. (8 ed.). Academic Press, Inc. p. 18. ISBN 978-0-12-384933-5. LCCN 2014010276. (Several previous editions as well.)
  2. ^ Niven, Ivan (October 1969). "Formal Power Series". American Mathematical Monthly. 76 (8): 871–889. doi:10.1080/00029890.1969.12000359.
  3. ^ Finkel, Hal (2010-07-13). "The differential transformation method and Miller's recurrence". arXiv:1007.2178 [math.CA].
  4. ^ a b Gould, H. W. (1974). "Coefficient Identities for Powers of Taylor and Dirichlet Series". The American Mathematical Monthly. 81 (1): 3–14. doi:10.2307/2318904. ISSN 0002-9890. JSTOR 2318904.
  5. ^ Zeilberger, Doron (1995). "The J.C.P. miller recurrence for exponentiating a polynomial, and its q-analog". Journal of Difference Equations and Applications. 1 (1): 57–60. doi:10.1080/10236199508808006 – via Taylor & Francis Online.
  6. ^ Richard, Stanley (2012). Enumerative combinatorics. Volume 1. Cambridge Stud. Adv. Math. Vol. 49. Cambridge: Cambridge University Press. ISBN 978-1-107-60262-5. MR 2868112.
  7. ^ Ira, Gessel (2016), "Lagrange inversion", Journal of Combinatorial Theory, Series A, 144: 212–249, arXiv:1609.05988, doi:10.1016/j.jcta.2016.06.018, MR 3534068
  8. ^ Surya, Erlang; Warnke, Lutz (2023), "Lagrange Inversion Formula by Induction", The American Mathematical Monthly, 130 (10): 944–948, arXiv:2305.17576, doi:10.1080/00029890.2023.2251344, MR 4669236
  9. ^ Koch, Helmut (1997). Algebraic Number Theory. Encycl. Math. Sci. Vol. 62 (2nd printing of 1st ed.). Springer-Verlag. p. 167. ISBN 978-3-540-63003-6. Zbl 0819.11044.
  10. ^ Moran, Siegfried (1983). The Mathematical Theory of Knots and Braids: An Introduction. North-Holland Mathematics Studies. Vol. 82. Elsevier. p. 211. ISBN 978-0-444-86714-8. Zbl 0528.57001.
  11. ^ Droste, M., & Kuich, W. (2009). Semirings and Formal Power Series. Handbook of Weighted Automata, 3–28. doi:10.1007/978-3-642-01492-5_1, p. 12
  12. ^ Shamseddine, Khodr; Berz, Martin (2010). "Analysis on the Levi-Civita Field: A Brief Overview" (PDF). Contemporary Mathematics. 508: 215–237. doi:10.1090/conm/508/10002. ISBN 9780821847404.

Further reading

[edit]
  • W. Kuich. Semirings and formal power series: Their relevance to formal languages and automata theory. In G. Rozenberg and A. Salomaa, editors, Handbook of Formal Languages, volume 1, Chapter 9, pages 609–677. Springer, Berlin, 1997, ISBN 3-540-60420-0
  • Droste, M., & Kuich, W. (2009). Semirings and Formal Power Series. Handbook of Weighted Automata, 3–28. doi:10.1007/978-3-642-01492-5_1
  • Arto Salomaa (1990). "Formal Languages and Power Series". In Jan van Leeuwen (ed.). Formal Models and Semantics. Handbook of Theoretical Computer Science. Vol. B. Elsevier. pp. 103–132. ISBN 0-444-88074-7.