Real Analysis

Introduction

Real analysis is the rigorous study of the real numbers, limits, continuity, differentiation, integration, sequences, series, and the structural ideas that make calculus work. In this university-level aiMOOC, you move from computational familiarity with calculus to proof-based reasoning. You will learn not only how to calculate a limit or derivative, but also how to prove that a limit exists, explain why continuity matters, identify the role of completeness, and recognize where familiar calculus statements can fail.

The central theme is control of approximation. Real analysis turns intuitive phrases such as “arbitrarily close,” “eventually,” and “without gaps” into precise definitions. These definitions let you prove theorems that support calculus and prepare you for Measure theory, Functional analysis, Probability theory, Differential equations, Topology, and advanced numerical mathematics.

The video above is an MIT OpenCourseWare introduction to real numbers and proof-based real analysis. As you watch, notice how definitions and proofs replace reliance on graphs or numerical evidence alone.

Prerequisites and Learning Goals

You should be comfortable with single-variable Calculus, elementary algebra, functions, inequalities, and basic set notation. Prior proof experience is helpful but not required.

By the end of this aiMOOC, you should be able to:

Write rigorous proofs using definitions, quantifiers, contradiction, contrapositive arguments, induction, and counterexamples.
Use the order and completeness properties of the real numbers to prove existence statements.
Analyze sequences and series using precise convergence criteria.
Prove limits with epsilon-delta and sequential arguments.
Distinguish continuity and uniform continuity and apply compactness-based theorems.
Use compactness and connectedness to understand global properties of continuous functions.
Justify differentiation theorems such as Rolle's theorem and the Mean Value Theorem.
Define and analyze Riemann integration and explain the Fundamental Theorem of Calculus.

The Real Numbers and Completeness

The real numbers form an ordered field, but the property that makes them especially powerful for analysis is completeness. One common form is the least-upper-bound property: every nonempty set of real numbers that is bounded above has a least upper bound, also called its supremum.

If a set $A \subseteq ℝ$ is bounded above, a number $M$ is an upper bound when $a \leq M$ for every $a \in A$ . The supremum $s = sup A$ is an upper bound that is no larger than any other upper bound. The definition has two parts: every element of $A$ is at most $s$ , and for every $ε > 0$ there is some $a \in A$ with $s - ε < a$ . The second part expresses that no smaller number can still be an upper bound.

The rational numbers do not have this completeness property. For example, the rational numbers whose squares are less than 2 are bounded above in $ℚ$ , but they have no rational least upper bound. In $ℝ$ , the missing boundary point is $\sqrt{2}$ .

The image shows a Dedekind cut associated with $\sqrt{2}$ . It illustrates one way to construct the real numbers by filling gaps in the rationals.

Completeness appears in several equivalent or closely related forms in elementary real analysis. Important consequences include the monotone convergence theorem for sequences, the nested interval property, the Bolzano-Weierstrass theorem, and the fact that every Cauchy sequence of real numbers converges to a real number.

This MIT OpenCourseWare lecture develops the ordered-field structure and characterization of the real numbers.

A Proof Pattern with Suprema

Suppose $(a_{n})$ is increasing and bounded above. Let $L = sup {a_{n} : n \in ℕ}$ . To prove $a_{n} \to L$ , start with an arbitrary $ε > 0$ . Because $L - ε$ cannot be an upper bound, some term $a_{N}$ satisfies $L - ε < a_{N} \leq L$ . Since the sequence is increasing, every $n \geq N$ satisfies $L - ε < a_{N} \leq a_{n} \leq L$ . Hence $| a_{n} - L | < ε$ for all sufficiently large $n$ .

This proof is a model for real analysis: an abstract existence principle is converted into an explicit epsilon estimate.

Sequences, Cauchy Sequences, and Series

A sequence $(a_{n})$ converges to $L$ if for every $ε > 0$ there exists $N \in ℕ$ such that $n \geq N$ implies $| a_{n} - L | < ε$ . The index $N$ may depend on $ε$ , but it must work for all later terms.

The picture visualizes a convergent sequence whose terms approach zero. A graph can suggest convergence, but a proof must control all sufficiently large indices.

A sequence is a Cauchy sequence if its terms eventually become arbitrarily close to one another: for every $ε > 0$ there is an $N$ such that $m, n \geq N$ implies $| a_{m} - a_{n} | < ε$ . Every convergent sequence is Cauchy. The completeness of $ℝ$ gives the converse: every Cauchy sequence of real numbers converges.

This lecture introduces Cauchy sequences, completeness, and the transition from sequences to infinite series.

Subsequences and Bolzano-Weierstrass

A subsequence keeps infinitely many terms of a sequence in their original order. Subsequence arguments are fundamental because they isolate convergent behavior that may be hidden inside a nonconvergent sequence.

The Bolzano-Weierstrass theorem states that every bounded sequence of real numbers has a convergent subsequence. A common proof repeatedly bisects a bounded interval and chooses a half containing infinitely many sequence terms. The nested intervals shrink to a point, and a carefully selected subsequence converges to that point.

This interval-halving idea later reappears in compactness arguments, root-finding methods, and multivariable analysis.

Infinite Series

An infinite series $\sum_{n = 1}^{\infty} a_{n}$ converges when its sequence of partial sums $s_{N} = \sum_{n = 1}^{N} a_{n}$ converges. Therefore, every series question is ultimately a sequence question.

Useful convergence tools include the comparison test, limit comparison test, ratio test, root test, alternating-series test, and Cauchy criterion. Remember that $a_{n} \to 0$ is necessary for convergence of $\sum a_{n}$ , but it is not sufficient. The harmonic series is the standard counterexample.

Absolute convergence, conditional convergence, and rearrangement show that infinite addition behaves differently from finite addition. These distinctions become even more important when you study power series, Fourier series, and function spaces.

Limits and Continuity

For a function $f$ , the statement $lim_{x \to a} f (x) = L$ means:

$\forall ε > 0 \exists δ > 0 \forall x, 0 < | x - a | < δ \Rightarrow | f (x) - L | < ε .$

The order of the quantifiers matters. You must choose $δ$ after seeing $ε$ , and that one choice of $δ$ must work for every allowed $x$ near $a$ .

An epsilon-delta proof typically has two phases. In the discovery phase, you work backward from $| f (x) - L | < ε$ to guess a useful restriction on $| x - a |$ . In the formal phase, you present a forward proof: given $ε > 0$ , choose a specific $δ > 0$ , assume $0 < | x - a | < δ$ , and derive the desired epsilon inequality.

Sequential Criterion for Limits

A powerful equivalent formulation says that $lim_{x \to a} f (x) = L$ if and only if for every sequence $(x_{n})$ with $x_{n} \neq a$ and $x_{n} \to a$ , the sequence $f (x_{n})$ converges to $L$ . This criterion is especially useful for disproving a proposed limit: find two sequences approaching the same point whose function values have different limiting behavior.

This MIT OpenCourseWare lecture connects limits of functions, limits of sequences, and continuity.

Continuity and Uniform Continuity

A function is continuous at $a$ when $lim_{x \to a} f (x) = f (a)$ . Equivalently, for every $ε > 0$ there is a $δ > 0$ such that $| x - a | < δ$ implies $| f (x) - f (a) | < ε$ .

For ordinary continuity, the allowable $δ$ may depend on both $ε$ and the point $a$ . For uniform continuity on a set $E$ , one $δ$ must work everywhere in the domain for a given $ε$ :

$\forall ε > 0 \exists δ > 0 \forall x, y \in E, | x - y | < δ \Rightarrow | f (x) - f (y) | < ε .$

The Heine-Cantor theorem says that every continuous real-valued function on a compact set is uniformly continuous. On a closed bounded interval, this gives a strong global form of continuity.

This lecture develops uniform continuity and connects it with differentiation.

Compactness, Connectedness, and Global Theorems

In $ℝ$ , the Heine-Borel theorem states that a set is compact if and only if it is closed and bounded. Compactness has several equivalent formulations in this setting. Every open cover has a finite subcover, and every sequence in a compact set has a convergent subsequence whose limit remains in the set.

Compactness turns local information into global conclusions. If $f : [a, b] \to ℝ$ is continuous, then:

the Extreme value theorem guarantees that $f$ attains both a maximum and a minimum;
the Heine–Cantor theorem guarantees that $f$ is uniformly continuous;
compactness supplies convergent subsequences in many existence proofs.

A set in $ℝ$ is connected precisely when it is an interval. The continuous image of a connected set is connected. This structural fact leads to the Intermediate Value Theorem: if $f$ is continuous on $[a, b]$ and a number $y$ lies between $f (a)$ and $f (b)$ , then some $c \in [a, b]$ satisfies $f (c) = y$ .

The theorem proves existence without necessarily producing a closed-form solution. It justifies methods such as bisection for finding roots.

The Cantor Set as a Stress Test

The Cantor set is constructed from $[0, 1]$ by repeatedly removing the open middle third from every remaining interval.

The Cantor set is closed, bounded, compact, perfect, nowhere dense, uncountable, and has length zero in the sense of Lebesgue measure. It is a valuable example because it challenges geometric intuition: a set can contain uncountably many points while containing no interval. Studying the Cantor set prepares you for Topology, Measure theory, fractals, and more advanced real analysis.

Differentiation and the Mean Value Theorem

The derivative of $f$ at $a$ is the limit

$f^{'} (a) = lim_{h \to 0} \frac{f (a + h) - f (a)}{h},$

provided the limit exists. Differentiability implies continuity, but continuity does not imply differentiability.

The Mean Value Theorem states that if $f$ is continuous on $[a, b]$ and differentiable on $(a, b)$ , then there is some $c \in (a, b)$ such that

$f^{'} (c) = \frac{f (b) - f (a)}{b - a} .$

This theorem connects local information about derivatives with global changes in function values. It yields important consequences: a function with zero derivative on an interval is constant; derivative bounds give Lipschitz-type estimates; and monotonicity can be deduced from the sign of the derivative.

Rolle's theorem is the special case in which $f (a) = f (b)$ . It is frequently used as the core lemma in proofs of the Mean Value Theorem and related results.

Riemann Integration

The Riemann integral makes the intuitive idea of area and accumulation rigorous by approximating a function with finite sums over partitions.

For a bounded function $f : [a, b] \to ℝ$ , choose a partition $P$ of $[a, b]$ . On each subinterval, take the infimum and supremum of $f$ . The resulting lower and upper sums bound any reasonable rectangle-based approximation. A bounded function is Riemann integrable when its upper and lower integrals agree. Equivalently, for every $ε > 0$ there is a partition for which the upper sum minus the lower sum is less than $ε$ .

Every continuous function on a closed bounded interval is Riemann integrable. Every monotone function on such an interval is also Riemann integrable. These results depend on the ability to control oscillation on sufficiently small subintervals.

The Fundamental Theorem of Calculus links differentiation and integration. In one standard form, if $f$ is continuous on $[a, b]$ and $F (x) = \int_{a}^{x} f (t) d t$ , then $F$ is differentiable on $(a, b)$ and $F^{'} (x) = f (x)$ . Another form evaluates definite integrals using antiderivatives.

This MIT OpenCourseWare lecture proves the Fundamental Theorem of Calculus and related properties of the Riemann integral.

Proof Strategy in Real Analysis

A successful proof begins by identifying the exact definition that controls the statement. Before doing algebra, rewrite the goal in quantifier form and decide what must be chosen and what is given.

Common proof patterns include direct proof, proof by contrapositive, contradiction, induction, construction of subsequences, epsilon estimates, and counterexamples. In analysis, counterexamples are not side notes; they tell you why hypotheses are necessary.

For example, if a theorem assumes compactness, test what happens on an open or unbounded set. If a theorem assumes uniform convergence, search for a pointwise-convergent sequence that loses continuity in the limit. If a theorem assumes differentiability, test a function with a corner or cusp.

A useful proof-writing checklist is:

State the arbitrary objects and hypotheses clearly.
Name the definition or theorem you are using.
Make every dependence explicit, especially choices of $N$ or $δ$ .
Justify inequalities and limiting steps.
End by matching your conclusion to the exact statement you intended to prove.

Interactive Tasks

Quiz: Test Your Knowledge

Which property distinguishes the real numbers from the rational numbers in basic real analysis? (The least upper bound property) (!Closure under addition) (!Existence of additive inverses) (!Order compatibility with multiplication)

What must be true for a sequence to converge to a real number L? (Eventually every term lies within every positive error tolerance of L) (!Every term equals L after a finite index) (!The sequence must be monotone) (!The sequence must contain only positive terms)

What does completeness of the real numbers imply about Cauchy sequences? (Every Cauchy sequence converges to a real number) (!Every bounded sequence is monotone) (!Every divergent sequence is unbounded) (!Every series with small terms converges)

Which statement is the Bolzano-Weierstrass theorem for real sequences? (Every bounded sequence has a convergent subsequence) (!Every convergent sequence is strictly increasing) (!Every unbounded sequence has a finite limit) (!Every subsequence has the same first term)

In an epsilon-delta proof of a function limit, what may delta depend on? (The chosen positive epsilon) (!The particular nearby input point) (!The value of every later sequence term) (!The final index of a finite list)

What extra feature distinguishes uniform continuity from ordinary continuity? (One delta works across the whole domain for a fixed epsilon) (!The function must be differentiable everywhere) (!The domain must contain only integers) (!The function must have a constant derivative)

Which condition characterizes compact subsets of the real line? (Closed and bounded) (!Open and bounded) (!Closed and unbounded) (!Dense and countable)

What does the Intermediate Value Theorem guarantee for a continuous function on an interval? (Every value between the endpoint values is attained) (!Every value has exactly one preimage) (!The derivative exists at every point) (!The function is necessarily monotone)

Which statement follows from the Mean Value Theorem? (A function with zero derivative on an interval is constant) (!Every continuous function is differentiable) (!Every bounded function has zero derivative somewhere) (!Every differentiable function is periodic)

How is convergence of an infinite series defined? (The sequence of partial sums converges) (!The individual terms are all nonzero) (!The terms alternate in sign) (!The sequence of terms is bounded above)

Memory Game

Supremum	Least upper bound of a set
Cauchy sequence	Sequence whose terms become arbitrarily close to each other
Compactness	Property equivalent to being closed and bounded in the real line
Uniform continuity	Continuity with one distance tolerance working across the domain
Subsequence	Sequence formed by retaining terms in their original order
Riemann integral	Integral defined through limits of finite partition sums
Connectedness	Property that characterizes intervals in the real line
Counterexample	Example that disproves a universal claim

Drag and Drop

Match the correct terms.	Topic
Completeness	Every Cauchy sequence of real numbers converges
Bolzano-Weierstrass theorem	Every bounded real sequence has a convergent subsequence
Heine-Cantor theorem	Continuous functions on compact sets are uniformly continuous
Intermediate Value Theorem	Continuous functions on intervals attain intermediate values
Mean Value Theorem	A derivative equals an average rate of change at some interior point

...

Crossword Puzzle

Supremum	What is the least upper bound of a set called?
Cauchy	Which kind of sequence has terms that become arbitrarily close to each other?
Compactness	Which property is equivalent to closedness and boundedness for subsets of the real line?
Continuity	Which property means small input changes produce controlled output changes near a point?
Derivative	What object represents the limiting local rate of change of a function?
Integrable	What adjective describes a bounded function whose Riemann upper and lower integrals agree?

LearningApps

Cloze Text

The real numbers are distinguished from the rationals by

. A convergent sequence must eventually stay arbitrarily close to its

. A sequence whose terms become arbitrarily close to one another is called

. The epsilon-delta definition gives a precise meaning to a function

. A continuous function on a compact set is

. In the real line, a set is compact exactly when it is closed and

. The Mean Value Theorem connects a derivative with an average rate of

. The Riemann integral is built from finite sums over

.

Open-Ended Tasks

Easy

Proof vocabulary: Create a one-page glossary that explains implication, converse, contrapositive, contradiction, counterexample, necessary condition, and sufficient condition in your own words, with one original mathematical example for each term.
Sequence experiment: Use a spreadsheet, graphing tool, or short program to plot the first fifty terms of three sequences, then write a paragraph explaining which appear to converge and what a rigorous proof would still need to establish.
Epsilon neighborhood: Draw or digitally design a visual explanation of sequence convergence that marks an epsilon-neighborhood of a proposed limit and highlights the index after which all terms remain inside it.
Continuity interview: Interview a classmate about the difference between continuity and uniform continuity, then revise the explanation until it correctly identifies what the delta may depend on.

Standard

Epsilon-delta proof: Produce a complete epsilon-delta proof for a linear or quadratic function limit, and annotate each line to explain why the chosen delta works.
Counterexample collection: Build a mini-gallery of four counterexamples showing that a weakened hypothesis can break a familiar theorem, including one example about continuity, one about compactness, one about series, and one about differentiability.
Bolzano-Weierstrass visualization: Create a diagram or short video that demonstrates interval bisection for a bounded sequence and explains how an infinite subsequence is selected.
Riemann sum investigation: Approximate the integral of a continuous function numerically using left, right, and midpoint sums for increasingly fine partitions, then compare the numerical evidence with the definition of Riemann integrability.

Advanced

Cantor set project: Generate several stages of the Cantor set, estimate the total removed length, and write a proof-oriented explanation of why the remaining set is uncountable but contains no interval.
Uniform convergence: Investigate a sequence of continuous functions that converges pointwise but not uniformly, create a graph or animation, and explain which property of the limit exposes the failure of uniform convergence.
Compactness proof comparison: Write two proofs of one theorem on a closed bounded interval, one using sequences and one using open covers or nested intervals, then compare the logical structure and assumptions.
Analysis in practice: Interview a researcher, engineer, economist, physicist, or data scientist about where limiting arguments or error bounds enter their work, then connect the interview to at least two rigorous ideas from this course.

Learning Assessment

Proof of convergence: Prove that a specified monotone bounded sequence converges, and identify exactly where completeness of the real numbers enters the argument.
Limit transfer: Given an epsilon-delta limit statement, translate it into a sequential statement, prove one direction of the equivalence, and explain how the sequential version can be used to disprove a false limit.
Compactness application: Prove that a continuous function on a closed bounded interval attains a maximum, then construct an example showing why the conclusion can fail when compactness is removed.
Uniform continuity reasoning: Determine whether several functions on different domains are uniformly continuous, justify each decision rigorously, and distinguish local continuity arguments from global estimates.
Mean Value Theorem transfer: Use the Mean Value Theorem to derive a nontrivial inequality or error estimate, and explain why each hypothesis of the theorem is needed.
Riemann integrability argument: Show that a given bounded function is Riemann integrable or not integrable by controlling upper and lower sums, and connect the proof to oscillation on small intervals.

Evidence of Learning

A strong portfolio for this aiMOOC should show four kinds of evidence.

Area	Evidence
Knowledge	You can state and interpret core definitions and theorems involving completeness, convergence, continuity, compactness, differentiation, and Riemann integration.
Proof skills	You can unpack quantifiers, choose useful estimates, construct subsequences, use contradiction and contrapositive arguments, and test claims with counterexamples.
Products	You can produce rigorous written proofs, annotated diagrams, computational experiments, explanatory videos, or short mathematical reports that connect intuition with definitions.
Transfer	You can recognize when a new problem is controlled by completeness, compactness, continuity, convergence, or an error estimate, and select an appropriate theorem or proof strategy.

OERs on the Topic

The English Wikipedia article on Real analysis provides a broad reference map for the subject and links to related areas.

For deeper study, the MIT OpenCourseWare lectures embedded throughout this aiMOOC provide university-level explanations and proofs. You can use them to compare proof styles, revisit difficult definitions, and see how the topics fit into a full real-analysis course.

Linked Learning Areas

Real analysis links the logic of proof with the rigorous foundations of calculus. Its central ideas are the completeness of the real numbers, convergence of sequences and series, epsilon-delta limits, continuity and uniform continuity, compactness and connectedness, differentiation, and Riemann integration. These ideas form a bridge to topology, measure theory, probability, differential equations, numerical analysis, optimization, and functional analysis.

Real Analysis

aiMOOC Projects

Diesen aiMOOC als Battle spielen: Mathematik, Klasse 5-6, G-Niveau

Inhalte werden geladen ...

Mediathek wird aus dem Wiki geladen ...

Frag mich zum aktuellen Lerninhalt oder wähle eine kurze Hilfe.

Textmodus bereit

Sprache

English:Real Analysis