Elementary function: Difference between revisions

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
imported>MandrewV
Consolidate two citations of same source.
 
imported>Andy02124
 
Line 1: Line 1:
{{short description|A kind of mathematical function}}
{{Short description|Type of mathematical function}}
{{About| |the complexity class | Elementary recursive function}}
{{distinguish|Elementary recursive function}}
In [[mathematics]], an '''elementary function''' is a [[function (mathematics)|function]] of a single [[variable (mathematics)|variable]] ([[Function of a real variable|real]] or [[Complex analysis#Complex functions|complex]]) that is typically encountered by beginners.  The basic elementary functions are [[polynomial function]]s, [[rational function]]s, the [[trigonometric function]]s, the [[exponential function|exponential]] and [[logarithm]] functions, the [[n-th root]], and the [[inverse trigonometric function]]s, as well as those functions obtained by [[addition]], [[multiplication]], [[division (mathematics)|division]], and [[function composition|composition]] of these.  Some functions which are encountered by beginners are ''not'' elementary, such as the [[absolute value]] function and [[piecewise-defined function]]s. More generally, in modern mathematics, elementary functions comprise the set of functions previously enumerated, all [[algebraic function]]s (not often encountered by beginners), and all functions obtained by [[roots of a polynomial]] whose coefficients are elementary.


In [[mathematics]], an '''elementary function''' is a [[function (mathematics)|function]] of a single [[variable (mathematics)|variable]] (typically [[Function of a real variable|real]] or [[Complex analysis#Complex functions|complex]]) that is defined as taking [[addition|sums]], [[multiplication|products]], [[algebraic function|roots]] and [[composition of functions|compositions]] of [[finite set|finitely]] many [[Polynomial#Polynomial functions|polynomial]], [[Rational function|rational]], [[Trigonometric functions|trigonometric]], [[Hyperbolic functions|hyperbolic]], and [[Exponential function|exponential]] functions, and their [[Inverse function|inverses]] (e.g., [[Inverse trigonometric functions|arcsin]], [[Natural logarithm|log]], or ''x''<sup>1/''n''</sup>).<ref name=":1">{{Cite book|title=Calculus|last=Spivak, Michael.|date=1994|publisher=Publish or Perish|isbn=0914098896|edition=3rd|location=Houston, Tex.|pages=363|oclc=31441929}}</ref>
This list of elementary functions was originally set forth by [[Joseph Liouville]] in 1833. A key property is that all elementary functions have [[derivative]]s of any order, which are also elementary, and can be [[algorithmically]] computed by applying the [[differentiation rules]] (or the rules for [[implicit differentiation]] in the case of roots). The [[Taylor series]] of an elementary function converges in a neighborhood of every point of its domain.  More generally, they are [[global analytic function]]s, defined (possibly with [[multivalued function|multiple values]], such as the elementary function <math>\sqrt z</math> or <math>\log z</math>) for every [[complex number|complex]] argument, except at [[isolated point]]s. In contrast, [[antiderivative]]s of elementary functions need not be elementary and is difficult to decide whether a specific elementary function has an elementary antiderivative.


All elementary functions are continuous on their [[Domain of a function|domains]].
[[Liouville's theorem (differential algebra)|Liouville's result]] is that, if an elementary function has an elementary antiderivative, then this antiderivative is a linear combination of logarithms, where the coefficients and the arguments of the logarithms are elementary functions involved, in some sense, in the definition of the function. More than 130 years later, [[Risch algorithm]], named after [[Robert Henry Risch]], is an algorithm to decide whether an elementary function has an elementary antiderivative, and, if it has, to compute this antiderivative. Despite dealing with elementary functions, the Risch algorithm is far from elementary; {{as of|2025|lc=y}}, it seems that no complete implementation is available.


Elementary functions were introduced by [[Joseph Liouville]] in a series of papers from 1833 to 1841.<ref>{{harvnb|Liouville|1833a}}.</ref><ref>{{harvnb|Liouville|1833b}}.</ref><ref>{{harvnb|Liouville|1833c}}.</ref> An [[abstract algebra|algebraic]] treatment of elementary functions was started by [[Joseph Fels Ritt]] in the 1930s.<ref>{{harvnb|Ritt|1950}}.</ref> Many textbooks and dictionaries do not give a precise definition of the elementary functions, and mathematicians differ on it.<ref name=":0">{{Cite journal |last1=Subbotin |first1=Igor Ya. |last2=Bilotskii |first2=N. N. |date=March 2008 |title=Algorithms and Fundamental Concepts of Calculus |url=https://assets.nu.edu/assets/resources/pageResources/Journal_of_Research_March081.pdf |journal=Journal of Research in Innovative Teaching |volume=1 |issue=1 |pages=82–94}}</ref>
In late-nineteenth-century analysis, elementary functions were often classified into successive kinds according to the number of independent integrations required for their definitionFunctions expressible without any integration—those generated from rational functions by algebraic operations together with exponentiation, logarithms, and circular or hyperbolic trigonometric functions—were said to be elementary functions of the first kind (in the sense of Liouville). Functions defined by a single integration of an algebraic function, such as the error function and the elliptic integrals, were elementary functions of the second kind; their inverses, the elliptic functions, were considered of the same order. Higher "kinds" (third, fourth, etc.) corresponded to multiple integrals of algebraic functions, giving rise to hyperelliptic and more general Abelian functions.{{sfn|Forsyth|1893}}
 
The essential point of the classification was that the class of elementary functions of any given kind be closed under the elementary operations—addition, multiplication, composition, and differentiation—so that differentiation never leads outside the same class, while integration may ascend to the next higher kind.


== Examples ==
== Examples ==
Line 12: Line 15:
=== Basic examples ===
=== Basic examples ===
Elementary functions of a single variable {{mvar|x}} include:
Elementary functions of a single variable {{mvar|x}} include:
* [[Constant function]]s: <math>2,\ \pi,\ e,</math> etc.
* [[Constant function]]s: <math>2,\ \pi,\ e,</math> the [[Euler–Mascheroni constant]], [[Apéry's constant]], [[Khinchin's constant]], etc.  Any constant real (or complex) number.
* [[Exponentiation#Rational_exponents|Rational powers of {{mvar|x}}]]: <math>x,\ x^2,\ \sqrt{x}\ (x^\frac{1}{2}),\ x^\frac{2}{3},</math> etc.
* [[Exponentiation|Powers of {{tmath|x}}]]: <math>x^\alpha=e^{\alpha\log x}</math> etc. (The exponent can be any real or complex constant.)
* [[Exponential function]]s: <math>e^x, \ a^x</math>
* [[Exponential function]]s: <math>\textstyle e^x,\quad a^x=e^{x\log a}</math>
* [[Logarithm]]s: <math>\log x, \ \log_a x</math>
* [[Logarithm]]s: <math>\textstyle \log x, \quad\log_a x=\frac {\log x}{\log a}</math>
* [[Trigonometric function]]s: <math>\sin x,\ \cos x,\ \tan x,</math> etc.
* [[Trigonometric function]]s: <math>\textstyle\sin x=\frac{e^{ix}-e^{-ix}}{2i},\ \cos x=\frac{e^{ix}+e^{-ix}}{2},\ \tan x=\frac{\sin x}{\cos x},\ </math> etc.
* [[Inverse trigonometric function]]s: <math>\arcsin x,\ \arccos x,</math> etc.
* [[Inverse trigonometric function]]s: <math>\arcsin x,\ \arccos x,</math> etc.
* [[Hyperbolic function]]s: <math>\sinh x,\ \cosh x,</math> etc.
* [[Hyperbolic function]]s: <math>\sinh x,\ \cosh x,</math> etc.
* [[Inverse hyperbolic function]]s: <math>\operatorname{arsinh} x,\ \operatorname{arcosh} x,</math> etc.
* [[Inverse hyperbolic function]]s: <math>\operatorname{arsinh} x,\ \operatorname{arcosh} x,</math> etc.
* All functions obtained by adding, subtracting, multiplying or dividing a finite number of any of the previous functions<ref>{{cite book|title=Ordinary Differential Equations|date=1985|publisher=Dover|isbn=0-486-64940-7|page=[https://archive.org/details/ordinarydifferen00tene_0/page/17 17]|url-access=registration|url=https://archive.org/details/ordinarydifferen00tene_0/page/17}}</ref>
* All functions obtained by adding, subtracting, multiplying or dividing a finite number of any of the previous functions<ref>{{cite book|author=Morris Tenenbaum|title=Ordinary Differential Equations|date=1985|publisher=Dover|isbn=0-486-64940-7|page=[https://archive.org/details/ordinarydifferen00tene_0/page/17 17]|url-access=registration|url=https://archive.org/details/ordinarydifferen00tene_0/page/17}}</ref>
* All functions obtained by root extraction of a polynomial with coefficients in elementary functions<ref name=":1" />
* All functions obtained as [[root of a polynomial|roots]] of a polynomial whose coefficients are elementary functions<ref name=":1">{{Cite book|title=Calculus|last=Spivak, Michael.|date=1994|publisher=Publish or Perish|isbn=0914098896|edition=3rd|location=Houston, Tex.|pages=363|oclc=31441929}}</ref><ref>Ritt, chapter 1</ref>
* All functions obtained by [[function composition|composing]] a finite number of any of the previously listed functions
* All functions obtained by [[function composition|composing]] a finite number of any of the previously listed functions


Line 37: Line 40:
The last function is equal to <math>\arccos x</math>, the [[Inverse_trigonometric_functions#Logarithmic_forms|inverse cosine]], in the entire [[complex plane]].
The last function is equal to <math>\arccos x</math>, the [[Inverse_trigonometric_functions#Logarithmic_forms|inverse cosine]], in the entire [[complex plane]].


All [[monomial]]s, [[polynomial]]s, [[rational function]]s and [[algebraic function]]s are elementary.  
All [[monomial]]s, [[polynomial]]s, [[rational function]]s and [[algebraic function]]s are elementary.
 
The [[Absolute value|absolute value function]], for real <math>x</math>, is also elementary<!-- {{cn|date=February 2021}} the proof is already provided here? --> as it can be expressed as the composition of a power and root of <math>x</math>: <math display="inline">|x|=\sqrt{x^2}</math>.{{dubious|date=June 2024}}


=== Non-elementary functions ===
=== Non-elementary functions ===
Many mathematicians exclude non-[[Analytic function|analytic functions]] such as the [[Absolute value|absolute value function]] or discontinuous functions such as the [[step function]],<ref>{{Cite journal |last=Risch |first=Robert H. |date=1979 |title=Algebraic Properties of the Elementary Functions of Analysis |url=https://www.jstor.org/stable/2373917 |journal=American Journal of Mathematics |volume=101 |issue=4 |pages=743–759 |doi=10.2307/2373917 |jstor=2373917 |issn=0002-9327|url-access=subscription }}</ref><ref name=":0" /> but others allow them. Some have proposed extending the set to include, for example, the [[Lambert W function]].<ref>{{Cite journal |last=Stewart |first=Seán |date=2005 |title=A new elementary function for our curricula? |url=https://files.eric.ed.gov/fulltext/EJ720055.pdf |journal=Australian Senior Mathematics Journal |volume=19 |issue=2 |pages=8–26}}</ref>
All elementary functions are [[Analytic function|analytic]] in the following sense: they can be extended to [[functions of a complex variable]] (possibly [[multivalued function|multivalued]]) that are analytic except at finitely many points of the [[complex plane]].<ref>{{Cite journal |last=Risch |first=Robert H. |date=1979 |title=Algebraic Properties of the Elementary Functions of Analysis |url=https://www.jstor.org/stable/2373917 |journal=[[American Journal of Mathematics]] |volume=101 |issue=4 |pages=743–759 |doi=10.2307/2373917 |jstor=2373917 |issn=0002-9327|url-access=subscription }}</ref> Thus nonanalytic functions such as the [[absolute value]] function are not elementary,<ref>Watson and Whittaker 1927, footnote to p 82. In the context of elementary functions, the function <math>y=f(x)</math> defined as the root of <math>y^2-x^2=0</math> is two-valued: <math>y=\pm x</math>.</ref> nor are most other [[piecewise-defined function]]s.


Some examples of functions that are ''not'' elementary:
Not every analytic function is elementary. In fact, most [[special function]]s are not elementary. Non-elementary functions include:
 
* [[tetration]]
* the [[gamma function]]
* the [[gamma function]]
* non-elementary [[Liouvillian function#Examples|Liouvillian functions]], including  
* non-elementary [[Liouvillian function#Examples|Liouvillian functions]], including  
** the [[exponential integral]] (''Ei''), [[logarithmic integral]] (''Li'' or ''li'') and [[Fresnel integral|Fresnel integrals]] (''S'' and ''C'').
** the [[exponential integral]] (''Ei'') [[logarithmic integral]] (''Li'' or ''li'') and [[Fresnel integral|Fresnel integrals]] (''S'' and ''C'')
** the [[error function]], <math>\mathrm{erf}(x)=\frac{2}{\sqrt{\pi}}\int_0^x e^{-t^2}\,dt,</math>  a fact that may not be immediately obvious, but can be proven using the [[Risch algorithm]].
** the [[error function]], <math>\mathrm{erf}(x)=\frac{2}{\sqrt{\pi}}\int_0^x e^{-t^2}\,dt,</math>  a fact that may not be immediately obvious, but can be proven using the [[Risch algorithm]]
* other [[nonelementary integral]]s, including the [[Dirichlet integral]] and [[elliptic integral]].
* other [[nonelementary integral]]s, including the [[Dirichlet integral]] and [[elliptic integral]]


== Closure ==
== Closure ==
It follows directly from the definition that the set of elementary functions is [[closure (mathematics)|closed]] under arithmetic operations, root extraction and composition. The elementary functions are closed under [[derivative|differentiation]]. They are not closed under [[series (mathematics)|limits and infinite sums]]. Importantly, the elementary functions are {{em|not}} closed under [[antiderivative|integration]], as shown by [[Liouville's theorem (differential algebra)|Liouville's theorem]], see [[nonelementary integral]]. The [[Liouvillian function]]s are defined as the elementary functions and, recursively, the integrals of the Liouvillian functions.
It follows directly from the definition that the set of elementary functions is [[closure (mathematics)|closed]] under arithmetic operations, (algebraic) root extraction and composition. The elementary functions are closed under [[derivative|differentiation]]. They are not closed under [[series (mathematics)|limits and infinite sums]]. Importantly, the elementary functions are {{em|not}} closed under [[antiderivative|integration]], as shown by [[Liouville's theorem (differential algebra)|Liouville's theorem]], see [[nonelementary integral]]. The [[Liouvillian function]]s are defined as the elementary functions and, recursively, the integrals of the Liouvillian functions.


==Differential algebra==
==Differential algebra==
Some have proposed extending the set of elementary functions by extending with certain [[transcendental function]]s, to include, for example, the [[Lambert W function]]<ref>{{Cite journal |last=Stewart |first=Seán |date=2005 |title=A new elementary function for our curricula? |url=https://files.eric.ed.gov/fulltext/EJ720055.pdf |journal=Australian Senior Mathematics Journal |volume=19 |issue=2 |pages=8–26}}</ref> or [[elliptic function]]s,<ref>Ince, E. L. (1956) [1926]. ''Ordinary Differential Equations''. New York: Dover Publications. ISBN 0-486-60339-4, footnote to p 330</ref> all of which are analytic. The key attribute, from the perspective of the Liouville theorem, is that as a class, they are closed under taking derivatives. For example, the Lambert function <math>w=W(z)</math>, which is defined implicitly by the equation <math>we^w=z</math>, has a derivative which can be obtained by [[implicit differentiation]]:
<math>W'(z) = \frac{e^{-W(z)}}{1+W(z)},</math>
which is again "elementary", provided that <math>W(z)</math> is.


The mathematical definition of an '''elementary function''', or a function in elementary form, is considered in the context of [[differential algebra]].  A differential algebra is an algebra with the extra operation of derivation (algebraic version of differentiation).  Using the derivation operation new equations can be written and their solutions used in [[field extension|extensions]] of the algebra.  By starting with the [[field (mathematics)|field]] of [[rational function]]s, two special types of transcendental extensions (the logarithm and the exponential) can be added to the field building a tower containing elementary functions.
The mathematical definition of an ''elementary function'' is formalized in [[differential algebra]].  A [[differential field]] is a [[field (mathematics)|field]]  with an extra operation of derivation (algebraic version of differentiation).  Using the derivation operation new equations can be written and their solutions used in [[field extension|extensions]] of the algebra.  By starting with the [[field (mathematics)|field]] of [[rational function]]s, two special types of transcendental extensions (the logarithm and the exponential) can be added to the field building a tower containing elementary functions.
 
A '''differential field''' ''F'' is a field ''F''<sub>0</sub> (rational functions over the [[rational number|rationals]] '''Q''' for example) together with a derivation map ''u''&nbsp;→&nbsp;∂''u''.  (Here ∂''u'' is a new function. Sometimes the notation ''u''&prime; is used.) The derivation captures the properties of differentiation, so that for any two elements of the base field, the derivation is linear
 
: <math>\partial (u + v) = \partial u + \partial v </math>


and satisfies the [[product rule|Leibniz product rule]]
A ''differential field'' {{tmath|F}} is a field  together with a [[derivation (differential algebra)|derivation]]  {{tmath|u\mapsto \partial u}} that maps {{tmath|F}}  to itself. The derivation generalizes [[derivative]], being linear (thaat is, {{tmath|1=\partial (u + v) = \partial u + \partial v}}) and satisfying the [[product rule|Leibniz product rule]] (that is,{{tmath|1=\partial(u\cdot v)=\partial u\cdot v+u\cdot\partial v}}) for every two elements {{tmath|u}} and {{tmath|v}} in {{tmath|F}}. The [[rational function]]s over {{tmath|\Q}} of {{tmath|\C}} form a basic examples of differential fields, when equipped with the usual derivative.


: <math>\partial(u\cdot v)=\partial u\cdot v+u\cdot\partial v\,.</math>
An element {{math|''h''}} of {{tmath|F}} is a constant if {{tmath|1=\partial h=0}}. The constants of {{tmath|F}} form a dfferential field with zero derivative. Care must be taken that a differential field extension of a differential field may enlarge the field of constants.


An element ''h'' is a constant if ''∂h&nbsp;=&nbsp;0''.  If the base field is over the rationals, care must be taken when extending the field to add the needed transcendental constants.
A function {{mvar|u}} of a differential extension {{mvar|G}} of a differential field {{mvar|F}} is an '''elementary function''' over {{mvar|F}} if it belongs to a finite chain (for inclusion) of differential subfields of {{mvar|G}} that starts from {{mvar|F}} and is such that each is generated over the preceding one by a function that is either
* [[Algebraic function|algebraic]] over the preceding field, or
* an ''exponential'', that is, {{tmath|1=\partial u = u\partial a}}  for some {{tmath|a\in F}}, or
* a ''logarithm'', that is, {{tmath|1=\partial u = \partial a/a}} for some {{tmath|a\in F}}.
(see [[Liouville's theorem (differential algebra)|Liouville's theorem]])


A function ''u'' of a differential extension ''F''[''u''] of a differential field ''F'' is an '''elementary function''' over ''F'' if the function ''u''
With this definition, the usual elementary functions are exactly the function that are elementary over the field of the [[rational function]]s. This generalized definition allows considering every transcendental function as elementary for applying Liouville's theorem.
* is [[Algebraic function|algebraic]] over ''F'', or
* is an '''exponential''', that is, ∂''u'' = ''u'' ∂''a''  for ''a'' ∈ ''F'', or
* is a '''logarithm''', that is, ∂''u'' = ∂''a''&nbsp;/&nbsp;a for ''a'' ∈ ''F''.
(see also [[Liouville's theorem (differential algebra)|Liouville's theorem]])


==See also==
==See also==


* {{annotated link|Algebraic function}}
* [[Algebraic function]]
* {{annotated link|Closed-form expression}}
* {{anl|Closed-form expression}}
* {{annotated link|Differential Galois theory}}
* [[Differential Galois theory]]
* {{annotated link|Elementary function arithmetic}}
* {{anl|Elementary function arithmetic}}
* {{annotated link|Liouville's theorem (differential algebra)}}
* {{anl|Liouville's theorem (differential algebra)}}
* {{annotated link|Tarski's high school algebra problem}}
* [[Tarski's high school algebra problem]]
* {{annotated link|Transcendental function}}
* {{anl|Transcendental function}}
* {{annotated link|Tupper's self-referential formula}}


==Notes==
==Notes==
Line 91: Line 88:


==References==
==References==
* {{cite book |last=Forsyth |first=Andrew |author-link=Andrew Forsyth |title=Theory of Functions of a Complex Variable |date=1893 |url=https://archive.org/details/theoryoffunction00fors/?q=region |publisher=Cambridge |jfm=25.0652.01 }}
*{{Cite journal
*{{Cite journal
  | last = Liouville
  | last = Liouville

Latest revision as of 02:18, 1 November 2025

Template:Short description Script error: No such module "Distinguish". In mathematics, an elementary function is a function of a single variable (real or complex) that is typically encountered by beginners. The basic elementary functions are polynomial functions, rational functions, the trigonometric functions, the exponential and logarithm functions, the n-th root, and the inverse trigonometric functions, as well as those functions obtained by addition, multiplication, division, and composition of these. Some functions which are encountered by beginners are not elementary, such as the absolute value function and piecewise-defined functions. More generally, in modern mathematics, elementary functions comprise the set of functions previously enumerated, all algebraic functions (not often encountered by beginners), and all functions obtained by roots of a polynomial whose coefficients are elementary.

This list of elementary functions was originally set forth by Joseph Liouville in 1833. A key property is that all elementary functions have derivatives of any order, which are also elementary, and can be algorithmically computed by applying the differentiation rules (or the rules for implicit differentiation in the case of roots). The Taylor series of an elementary function converges in a neighborhood of every point of its domain. More generally, they are global analytic functions, defined (possibly with multiple values, such as the elementary function z or logz) for every complex argument, except at isolated points. In contrast, antiderivatives of elementary functions need not be elementary and is difficult to decide whether a specific elementary function has an elementary antiderivative.

Liouville's result is that, if an elementary function has an elementary antiderivative, then this antiderivative is a linear combination of logarithms, where the coefficients and the arguments of the logarithms are elementary functions involved, in some sense, in the definition of the function. More than 130 years later, Risch algorithm, named after Robert Henry Risch, is an algorithm to decide whether an elementary function has an elementary antiderivative, and, if it has, to compute this antiderivative. Despite dealing with elementary functions, the Risch algorithm is far from elementary; Template:As of, it seems that no complete implementation is available.

In late-nineteenth-century analysis, elementary functions were often classified into successive kinds according to the number of independent integrations required for their definition. Functions expressible without any integration—those generated from rational functions by algebraic operations together with exponentiation, logarithms, and circular or hyperbolic trigonometric functions—were said to be elementary functions of the first kind (in the sense of Liouville). Functions defined by a single integration of an algebraic function, such as the error function and the elliptic integrals, were elementary functions of the second kind; their inverses, the elliptic functions, were considered of the same order. Higher "kinds" (third, fourth, etc.) corresponded to multiple integrals of algebraic functions, giving rise to hyperelliptic and more general Abelian functions.Template:Sfn

The essential point of the classification was that the class of elementary functions of any given kind be closed under the elementary operations—addition, multiplication, composition, and differentiation—so that differentiation never leads outside the same class, while integration may ascend to the next higher kind.

Examples

Basic examples

Elementary functions of a single variable Template:Mvar include:

Certain elementary functions of a single complex variable Template:Mvar, such as z and logz, may be multivalued. Additionally, certain classes of functions may be obtained by others using the final two rules. For example, the exponential function ez composed with addition, subtraction, and division provides the hyperbolic functions, while initial composition with iz instead provides the trigonometric functions.

Composite examples

Examples of elementary functions include:

The last function is equal to arccosx, the inverse cosine, in the entire complex plane.

All monomials, polynomials, rational functions and algebraic functions are elementary.

Non-elementary functions

All elementary functions are analytic in the following sense: they can be extended to functions of a complex variable (possibly multivalued) that are analytic except at finitely many points of the complex plane.[4] Thus nonanalytic functions such as the absolute value function are not elementary,[5] nor are most other piecewise-defined functions.

Not every analytic function is elementary. In fact, most special functions are not elementary. Non-elementary functions include:

Closure

It follows directly from the definition that the set of elementary functions is closed under arithmetic operations, (algebraic) root extraction and composition. The elementary functions are closed under differentiation. They are not closed under limits and infinite sums. Importantly, the elementary functions are Template:Em closed under integration, as shown by Liouville's theorem, see nonelementary integral. The Liouvillian functions are defined as the elementary functions and, recursively, the integrals of the Liouvillian functions.

Differential algebra

Some have proposed extending the set of elementary functions by extending with certain transcendental functions, to include, for example, the Lambert W function[6] or elliptic functions,[7] all of which are analytic. The key attribute, from the perspective of the Liouville theorem, is that as a class, they are closed under taking derivatives. For example, the Lambert function w=W(z), which is defined implicitly by the equation wew=z, has a derivative which can be obtained by implicit differentiation: W(z)=eW(z)1+W(z), which is again "elementary", provided that W(z) is.

The mathematical definition of an elementary function is formalized in differential algebra. A differential field is a field with an extra operation of derivation (algebraic version of differentiation). Using the derivation operation new equations can be written and their solutions used in extensions of the algebra. By starting with the field of rational functions, two special types of transcendental extensions (the logarithm and the exponential) can be added to the field building a tower containing elementary functions.

A differential field Template:Tmath is a field together with a derivation Template:Tmath that maps Template:Tmath to itself. The derivation generalizes derivative, being linear (thaat is, Template:Tmath) and satisfying the Leibniz product rule (that is,Template:Tmath) for every two elements Template:Tmath and Template:Tmath in Template:Tmath. The rational functions over Template:Tmath of Template:Tmath form a basic examples of differential fields, when equipped with the usual derivative.

An element Template:Math of Template:Tmath is a constant if Template:Tmath. The constants of Template:Tmath form a dfferential field with zero derivative. Care must be taken that a differential field extension of a differential field may enlarge the field of constants.

A function Template:Mvar of a differential extension Template:Mvar of a differential field Template:Mvar is an elementary function over Template:Mvar if it belongs to a finite chain (for inclusion) of differential subfields of Template:Mvar that starts from Template:Mvar and is such that each is generated over the preceding one by a function that is either

(see Liouville's theorem)

With this definition, the usual elementary functions are exactly the function that are elementary over the field of the rational functions. This generalized definition allows considering every transcendental function as elementary for applying Liouville's theorem.

See also

Notes

Template:Reflist

References

  • Script error: No such module "citation/CS1".
  • Script error: No such module "Citation/CS1".
  • Script error: No such module "Citation/CS1".
  • Script error: No such module "Citation/CS1".
  • Script error: No such module "citation/CS1".
  • Script error: No such module "Citation/CS1".

Further reading

  • Script error: No such module "citation/CS1".

External links

Template:Authority control

  1. Script error: No such module "citation/CS1".
  2. Script error: No such module "citation/CS1".
  3. Ritt, chapter 1
  4. Script error: No such module "Citation/CS1".
  5. Watson and Whittaker 1927, footnote to p 82. In the context of elementary functions, the function y=f(x) defined as the root of y2x2=0 is two-valued: y=±x.
  6. Script error: No such module "Citation/CS1".
  7. Ince, E. L. (1956) [1926]. Ordinary Differential Equations. New York: Dover Publications. ISBN 0-486-60339-4, footnote to p 330