Extreme Value Theorem
   HOME

TheInfoList



OR:

In
calculus Calculus, originally called infinitesimal calculus or "the calculus of infinitesimals", is the mathematical study of continuous change, in the same way that geometry is the study of shape, and algebra is the study of generalizations of arithm ...
, the extreme value theorem states that if a real-valued
function Function or functionality may refer to: Computing * Function key, a type of key on computer keyboards * Function model, a structured representation of processes in a system * Function object or functor or functionoid, a concept of object-oriente ...
f is
continuous Continuity or continuous may refer to: Mathematics * Continuity (mathematics), the opposing concept to discreteness; common examples include ** Continuous probability distribution or random variable in probability and statistics ** Continuous ...
on the closed interval ,b/math>, then f must attain a
maximum In mathematical analysis, the maxima and minima (the respective plurals of maximum and minimum) of a function, known collectively as extrema (the plural of extremum), are the largest and smallest value of the function, either within a given ran ...
and a
minimum In mathematical analysis, the maxima and minima (the respective plurals of maximum and minimum) of a function, known collectively as extrema (the plural of extremum), are the largest and smallest value of the function, either within a given ran ...
, each at least once. That is, there exist numbers c and d in ,b/math> such that: f(c) \ge f(x) \ge f(d)\quad \forall x\in ,b/math> The extreme value theorem is more specific than the related boundedness theorem, which states merely that a continuous function f on the closed interval ,b/math> is bounded on that interval; that is, there exist real numbers m and M such that: m \le f(x) \le M\quad \forall x \in , b This does not say that M and m are necessarily the maximum and minimum values of f on the interval ,b which is what the extreme value theorem stipulates must also be the case. The extreme value theorem is used to prove Rolle's theorem. In a formulation due to
Karl Weierstrass Karl Theodor Wilhelm Weierstrass (german: link=no, Weierstraß ; 31 October 1815 – 19 February 1897) was a German mathematician often cited as the "father of modern analysis". Despite leaving university without a degree, he studied mathematics ...
, this theorem states that a continuous function from a non-empty
compact space In mathematics, specifically general topology, compactness is a property that seeks to generalize the notion of a closed and bounded subset of Euclidean space by making precise the idea of a space having no "punctures" or "missing endpoints", i ...
to a
subset In mathematics, Set (mathematics), set ''A'' is a subset of a set ''B'' if all Element (mathematics), elements of ''A'' are also elements of ''B''; ''B'' is then a superset of ''A''. It is possible for ''A'' and ''B'' to be equal; if they are ...
of the
real number In mathematics, a real number is a number that can be used to measure a ''continuous'' one-dimensional quantity such as a distance, duration or temperature. Here, ''continuous'' means that values can have arbitrarily small variations. Every real ...
s attains a maximum and a minimum.


History

The extreme value theorem was originally proven by
Bernard Bolzano Bernard Bolzano (, ; ; ; born Bernardus Placidus Johann Gonzal Nepomuk Bolzano; 5 October 1781 – 18 December 1848) was a Bohemian mathematician, logician, philosopher, theologian and Catholic priest of Italian extraction, also known for his liber ...
in the 1830s in a work ''Function Theory'' but the work remained unpublished until 1930. Bolzano's proof consisted of showing that a continuous function on a closed interval was bounded, and then showing that the function attained a maximum and a minimum value. Both proofs involved what is known today as the Bolzano–Weierstrass theorem. The result was also discovered later by Weierstrass in 1860.


Functions to which the theorem does not apply

The following examples show why the function domain must be closed and bounded in order for the theorem to apply. Each fails to attain a maximum on the given interval. # f(x)=x defined over [0, \infty) is not bounded from above. # f(x)= \frac defined over [0, \infty) is bounded but does not attain its least upper bound 1. # f(x)= \frac defined over (0,1] is not bounded from above. # f(x) = 1-x defined over (0,1] is bounded but never attains its least upper bound 1. Defining f(0)=0 in the last two examples shows that both theorems require continuity on ,b/math>.


Generalization to metric and topological spaces

When moving from the real line \mathbb to
metric spaces In mathematics, a metric space is a set together with a notion of ''distance'' between its elements, usually called points. The distance is measured by a function called a metric or distance function. Metric spaces are the most general settin ...
and general
topological spaces In mathematics, a topological space is, roughly speaking, a geometrical space in which closeness is defined but cannot necessarily be measured by a numeric distance. More specifically, a topological space is a set whose elements are called point ...
, the appropriate generalization of a closed bounded interval is a
compact set In mathematics, specifically general topology, compactness is a property that seeks to generalize the notion of a closed and bounded subset of Euclidean space by making precise the idea of a space having no "punctures" or "missing endpoints", i ...
. A set K is said to be compact if it has the following property: from every collection of
open set In mathematics, open sets are a generalization of open intervals in the real line. In a metric space (a set along with a distance defined between any two points), open sets are the sets that, with every point , contain all points that are suf ...
s U_\alpha such that \bigcup U_\alpha \supset K, a finite subcollection U_,\ldots,U_can be chosen such that \bigcup_^n U_ \supset K. This is usually stated in short as "every open cover of K has a finite subcover". The Heine–Borel theorem asserts that a subset of the real line is compact if and only if it is both closed and bounded. Correspondingly, a metric space has the Heine–Borel property if every closed and bounded set is also compact. The concept of a continuous function can likewise be generalized. Given topological spaces V,\ W, a function f:V\to W is said to be continuous if for every open set U\subset W, f^(U)\subset V is also open. Given these definitions, continuous functions can be shown to preserve compactness: Theorem. ''If V,\ W are topological spaces, f:V\to W is a continuous function, and K\subset V is compact, then f(K)\subset W is also compact.'' In particular, if W = \mathbb, then this theorem implies that f(K) is closed and bounded for any compact set K, which in turn implies that f attains its
supremum In mathematics, the infimum (abbreviated inf; plural infima) of a subset S of a partially ordered set P is a greatest element in P that is less than or equal to each element of S, if such an element exists. Consequently, the term ''greatest l ...
and
infimum In mathematics, the infimum (abbreviated inf; plural infima) of a subset S of a partially ordered set P is a greatest element in P that is less than or equal to each element of S, if such an element exists. Consequently, the term ''greatest low ...
on any (nonempty) compact set K. Thus, we have the following generalization of the extreme value theorem: Theorem. ''If'' K ''is a compact set and'' f:K\to \mathbb ''is a continuous function, then'' f ''is bounded and there exist'' p,q\in K ''such that'' f(p)=\sup_ f(x) ''and'' f(q) = \inf_ f(x)''.'' Slightly more generally, this is also true for an upper semicontinuous function. (see compact space#Functions and compact spaces).


Proving the theorems

We look at the proof for the
upper bound In mathematics, particularly in order theory, an upper bound or majorant of a subset of some preordered set is an element of that is greater than or equal to every element of . Dually, a lower bound or minorant of is defined to be an eleme ...
and the maximum of f. By applying these results to the function -f, the existence of the lower bound and the result for the minimum of f follows. Also note that everything in the proof is done within the context of the
real numbers In mathematics, a real number is a number that can be used to measure a ''continuous'' one-dimensional quantity such as a distance, duration or temperature. Here, ''continuous'' means that values can have arbitrarily small variations. Every real ...
. We first prove the boundedness theorem, which is a step in the proof of the extreme value theorem. The basic steps involved in the proof of the extreme value theorem are: # Prove the boundedness theorem. # Find a sequence so that its
image An image is a visual representation of something. It can be two-dimensional, three-dimensional, or somehow otherwise feed into the visual system to convey information. An image can be an artifact, such as a photograph or other two-dimensiona ...
converges to the
supremum In mathematics, the infimum (abbreviated inf; plural infima) of a subset S of a partially ordered set P is a greatest element in P that is less than or equal to each element of S, if such an element exists. Consequently, the term ''greatest l ...
of f. # Show that there exists a
subsequence In mathematics, a subsequence of a given sequence is a sequence that can be derived from the given sequence by deleting some or no elements without changing the order of the remaining elements. For example, the sequence \langle A,B,D \rangle is a ...
that converges to a point in the
domain Domain may refer to: Mathematics *Domain of a function, the set of input values for which the (total) function is defined **Domain of definition of a partial function **Natural domain of a partial function **Domain of holomorphy of a function * Do ...
. # Use continuity to show that the image of the subsequence converges to the supremum.


Proof of the boundedness theorem

Statement   If f(x) is continuous on ,b/math> then it is bounded on ,b/math> Suppose the function f is not bounded above on the interval ,b/math>. Then, for every natural number n, there exists an x_n \in ,b/math> such that f(x_n)>n. This defines a
sequence In mathematics, a sequence is an enumerated collection of objects in which repetitions are allowed and order matters. Like a set, it contains members (also called ''elements'', or ''terms''). The number of elements (possibly infinite) is calle ...
(x_n)_. Because ,b/math> is bounded, the
Bolzano–Weierstrass theorem In mathematics, specifically in real analysis, the Bolzano–Weierstrass theorem, named after Bernard Bolzano and Karl Weierstrass, is a fundamental result about convergence in a finite-dimensional Euclidean space \R^n. The theorem states that each ...
implies that there exists a convergent subsequence (x_)_ of (). Denote its limit by x. As ,b/math> is closed, it contains x. Because f is continuous at x, we know that f(x_) converges to the real number f(x) (as f is
sequentially continuous In topology and related fields of mathematics, a sequential space is a topological space whose topology can be completely characterized by its convergent/divergent sequences. They can be thought of as spaces that satisfy a very weak axiom of counta ...
at x). But f(x_) > n_k \geq k for every k, which implies that f(x_) diverges to + \infty , a contradiction. Therefore, f is bounded above on ,b/math>.  \Box


Alternative proof

Statement   If f(x) is continuous on ,b/math> then it is bounded on ,b/math> Proof    Consider the set B of points p in ,b/math> such that f(x) is bounded on ,p/math>. We note that a is one such point, for f(x) is bounded on ,a/math> by the value f(a). If e > a is another point, then all points between a and e also belong to B. In other words B is an interval closed at its left end by a. Now f is continuous on the right at a, hence there exists \delta>0 such that , f(x) - f(a), < 1 for all x in ,a+\delta/math>. Thus f is bounded by f(a) - 1 and f(a)+1 on the interval ,a+\delta/math> so that all these points belong to B. So far, we know that B is an interval of non-zero length, closed at its left end by a. Next, B is bounded above by b. Hence the set B has a supremum in ,b/math> ; let us call it s. From the non-zero length of B we can deduce that s > a. Suppose s. Now f is continuous at s, hence there exists \delta>0 such that , f(x) - f(s), < 1 for all x in -\delta,s+\delta/math> so that f is bounded on this interval. But it follows from the supremacy of s that there exists a point belonging to B, e say, which is greater than s-\delta/2. Thus f is bounded on ,e/math> which overlaps -\delta,s+\delta/math> so that f is bounded on ,s+\delta/math>. This however contradicts the supremacy of s. We must therefore have s=b. Now f is continuous on the left at s, hence there exists \delta>0 such that , f(x) - f(s), < 1 for all x in -\delta,s/math> so that f is bounded on this interval. But it follows from the supremacy of s that there exists a point belonging to B, e say, which is greater than s-\delta/2. Thus f is bounded on ,e/math> which overlaps -\delta,s/math> so that f is bounded on ,s/math>.  


Proof of the extreme value theorem

By the boundedness theorem, ''f'' is bounded from above, hence, by the
Dedekind-complete In mathematics, the least-upper-bound property (sometimes called completeness or supremum property or l.u.b. property) is a fundamental property of the real numbers. More generally, a partially ordered set has the least-upper-bound property if eve ...
ness of the real numbers, the least upper bound (supremum) ''M'' of ''f'' exists. It is necessary to find a point ''d'' in 'a'', ''b''such that ''M'' = ''f''(''d''). Let ''n'' be a natural number. As ''M'' is the ''least'' upper bound, ''M'' – 1/''n'' is not an upper bound for ''f''. Therefore, there exists ''dn'' in 'a'', ''b''so that ''M'' – 1/''n'' < ''f''(''dn''). This defines a sequence . Since ''M'' is an upper bound for ''f'', we have ''M'' – 1/''n'' < ''f''(''dn'') ≤ ''M'' for all ''n''. Therefore, the sequence converges to ''M''. The
Bolzano–Weierstrass theorem In mathematics, specifically in real analysis, the Bolzano–Weierstrass theorem, named after Bernard Bolzano and Karl Weierstrass, is a fundamental result about convergence in a finite-dimensional Euclidean space \R^n. The theorem states that each ...
tells us that there exists a subsequence , which converges to some ''d'' and, as 'a'', ''b''is closed, ''d'' is in 'a'', ''b'' Since ''f'' is continuous at ''d'', the sequence converges to ''f''(''d''). But is a subsequence of that converges to ''M'', so ''M'' = ''f''(''d''). Therefore, ''f'' attains its supremum ''M'' at ''d''. 


Alternative proof of the extreme value theorem

The set is a bounded set. Hence, its
least upper bound In mathematics, the infimum (abbreviated inf; plural infima) of a subset S of a partially ordered set P is a greatest element in P that is less than or equal to each element of S, if such an element exists. Consequently, the term ''greatest lo ...
exists by
least upper bound property In mathematics, the least-upper-bound property (sometimes called completeness or supremum property or l.u.b. property) is a fundamental property of the real numbers. More generally, a partially ordered set has the least-upper-bound property if eve ...
of the real numbers. Let  on . If there is no point ''x'' on 'a'', ''b''so that ''f''(''x'') = ''M'' ,then on 'a'', ''b'' Therefore, is continuous on 'a'', ''b'' However, to every positive number ''ε'', there is always some ''x'' in 'a'', ''b''such that because ''M'' is the least upper bound. Hence, , which means that is not bounded. Since every continuous function on a 'a'', ''b''is bounded, this contradicts the conclusion that was continuous on 'a'', ''b'' Therefore, there must be a point ''x'' in 'a'', ''b''such that ''f''(''x'') = ''M''.


Proof using the hyperreals

In the setting of
non-standard calculus In mathematics, nonstandard calculus is the modern application of infinitesimals, in the sense of nonstandard analysis, to infinitesimal calculus. It provides a rigorous justification for some arguments in calculus that were previously considered m ...
, let ''N''  be an infinite
hyperinteger In nonstandard analysis, a hyperinteger ''n'' is a hyperreal number that is equal to its own integer part. A hyperinteger may be either finite or infinite. A finite hyperinteger is an ordinary integer. An example of an infinite hyperinteger is g ...
. The interval , 1has a natural hyperreal extension. Consider its partition into ''N'' subintervals of equal
infinitesimal In mathematics, an infinitesimal number is a quantity that is closer to zero than any standard real number, but that is not zero. The word ''infinitesimal'' comes from a 17th-century Modern Latin coinage ''infinitesimus'', which originally referr ...
length 1/''N'', with partition points ''xi'' = ''i'' /''N'' as ''i'' "runs" from 0 to ''N''. The function ''ƒ''  is also naturally extended to a function ''ƒ''* defined on the hyperreals between 0 and 1. Note that in the standard setting (when ''N''  is finite), a point with the maximal value of ''ƒ'' can always be chosen among the ''N''+1 points ''xi'', by induction. Hence, by the
transfer principle In model theory, a transfer principle states that all statements of some language that are true for some structure are true for another structure. One of the first examples was the Lefschetz principle, which states that any sentence in the first- ...
, there is a hyperinteger ''i''0 such that 0 ≤ ''i''0 ≤ ''N'' and f^*(x_)\geq f^*(x_i)  for all ''i'' = 0, ..., ''N''. Consider the real point c = \mathbf(x_) where st is the
standard part function In nonstandard analysis, the standard part function is a function from the limited (finite) hyperreal numbers to the real numbers. Briefly, the standard part function "rounds off" a finite hyperreal to the nearest real. It associates to every suc ...
. An arbitrary real point ''x'' lies in a suitable sub-interval of the partition, namely x\in _i,x_/math>, so that  st(''xi'') = ''x''. Applying st to the inequality f^*(x_)\geq f^*(x_i), we obtain \mathbf(f^*(x_))\geq \mathbf(f^*(x_i)). By continuity of ''ƒ''  we have :\mathbf(f^*(x_))= f(\mathbf(x_))=f(c). Hence ''ƒ''(''c'') ≥ ''ƒ''(''x''), for all real ''x'', proving ''c'' to be a maximum of ''ƒ''.


Proof from first principles

Statement      If f(x) is continuous on ,b/math> then it attains its supremum on ,b/math> Proof      By the Boundedness Theorem, f(x) is bounded above on ,b/math> and by the completeness property of the real numbers has a supremum in , b/math>. Let us call it M, or M ,b/math>. It is clear that the restriction of f to the subinterval ,x/math> where x\le b has a supremum M , x/math> which is less than or equal to M, and that M ,x/math> increases from f(a) to M as x increases from a to b. If f(a)=M then we are done. Suppose therefore that f(a) and let d=M-f(a). Consider the set L of points x in ,b/math> such that M ,x M. Clearly a\in L ; moreover if e>a is another point in L then all points between a and e also belong to L because M ,x/math> is monotonic increasing. Hence L is a non-empty interval, closed at its left end by a. Now f is continuous on the right at a, hence there exists \delta>0 such that , f(x)-f(a), < d/2 for all x in ,a+\delta/math>. Thus f is less than M-d/2 on the interval ,a+\delta/math> so that all these points belong to L. Next, L is bounded above by b and has therefore a supremum in ,b/math>: let us call it s. We see from the above that s > a. We will show that s is the point we are seeking i.e. the point where f attains its supremum, or in other words f(s)=M. Suppose the contrary viz. f(s). Let d=M-f(s) and consider the following two cases: # s.   As f is continuous at s, there exists \delta>0 such that , f(x)-f(s), < d/2 for all x in -\delta,s+\delta/math>. This means that f is less than M-d/2 on the interval -\delta,s+\delta/math>. But it follows from the supremacy of s that there exists a point, e say, belonging to L which is greater than s-\delta. By the definition of L, M ,e M. Let d_1=M-M ,e/math> then for all x in ,e/math>, f(x)\le M-d_1. Taking d_2 to be the minimum of d/2 and d_1, we have f(x)\le M-d_2 for all x in ,s+\delta/math>. Hence M ,s+\deltaM so that s+\delta \in L. This however contradicts the supremacy of s and completes the proof. # s=b.   As f is continuous on the left at s, there exists \delta>0 such that , f(x)-f(s), < d/2 for all x in -\delta,s/math>. This means that f is less than M-d/2 on the interval -\delta,s/math>. But it follows from the supremacy of s that there exists a point, e say, belonging to L which is greater than s-\delta. By the definition of L, M ,e M. Let d_1=M-M ,e/math> then for all x in ,e/math>, f(x)\le M-d_1. Taking d_2 to be the minimum of d/2 and d_1, we have f(x)\le M-d_2 for all x in ,b/math>. This contradicts the supremacy of M and completes the proof.


Extension to semi-continuous functions

If the continuity of the function ''f'' is weakened to
semi-continuity In mathematical analysis, semicontinuity (or semi-continuity) is a property of extended real-valued functions that is weaker than continuity. An extended real-valued function f is upper (respectively, lower) semicontinuous at a point x_0 if, rou ...
, then the corresponding half of the boundedness theorem and the extreme value theorem hold and the values –∞ or +∞, respectively, from the
extended real number line In mathematics, the affinely extended real number system is obtained from the real number system \R by adding two infinity elements: +\infty and -\infty, where the infinities are treated as actual numbers. It is useful in describing the algebra ...
can be allowed as possible values. More precisely: Theorem: If a function is upper semi-continuous, meaning that \limsup_ f(y) \le f(x) for all ''x'' in 'a'',''b'' then ''f'' is bounded above and attains its supremum. ''Proof:'' If ''f''(''x'') = –∞ for all ''x'' in 'a'',''b'' then the supremum is also –∞ and the theorem is true. In all other cases, the proof is a slight modification of the proofs given above. In the proof of the boundedness theorem, the upper semi-continuity of ''f'' at ''x'' only implies that the
limit superior In mathematics, the limit inferior and limit superior of a sequence can be thought of as limiting (that is, eventual and extreme) bounds on the sequence. They can be thought of in a similar fashion for a function (see limit of a function). For a ...
of the subsequence is bounded above by ''f''(''x'') < ∞, but that is enough to obtain the contradiction. In the proof of the extreme value theorem, upper semi-continuity of ''f'' at ''d'' implies that the limit superior of the subsequence is bounded above by ''f''(''d''), but this suffices to conclude that ''f''(''d'') = ''M''.  Applying this result to −''f'' proves: Theorem: If a function is lower semi-continuous, meaning that \liminf_ f(y) \geq f(x) for all ''x'' in 'a'',''b'' then ''f'' is bounded below and attains its
infimum In mathematics, the infimum (abbreviated inf; plural infima) of a subset S of a partially ordered set P is a greatest element in P that is less than or equal to each element of S, if such an element exists. Consequently, the term ''greatest low ...
. A real-valued function is upper as well as lower semi-continuous, if and only if it is continuous in the usual sense. Hence these two theorems imply the boundedness theorem and the extreme value theorem.


References


Further reading

* *


External links


A Proof for extreme value theorem
at
cut-the-knot Alexander Bogomolny (January 4, 1948 July 7, 2018) was a Soviet-born Israeli-American mathematician. He was Professor Emeritus of Mathematics at the University of Iowa, and formerly research fellow at the Moscow Institute of Electronics and Math ...

Extreme Value Theorem
by Jacqueline Wandzura with additional contributions by Stephen Wandzura, the
Wolfram Demonstrations Project The Wolfram Demonstrations Project is an organized, open-source collection of small (or medium-size) interactive programs called Demonstrations, which are meant to visually and interactively represent ideas from a range of fields. It is hos ...
. * *
Mizar system The Mizar system consists of a formal language for writing mathematical definitions and proofs, a proof assistant, which is able to mechanically check proofs written in this language, and a library of formalized mathematics, which can be used in ...
proof: http://mizar.org/version/current/html/weierstr.html#T15 {{Calculus topics Articles containing proofs Theorems in calculus Theorems in real analysis