Introduction to Real Analysis: Why Rigor Matters

Real analysis is the course where calculus stops being a set of techniques and becomes a set of proofs. Many university students find it a shock: suddenly "obvious" facts such as \( \frac{1}{n} \to 0 \) need careful justification. This guide explains why that rigor is necessary and shows you, step by step, how to write the two most important kinds of proof.

Why Calculus Needed Rescuing

Newton and Leibniz built calculus on intuitive ideas like "infinitely small quantities". It worked brilliantly, but it also produced puzzles that intuition could not settle:

• The series \( 1 - 1 + 1 - 1 + \cdots \). Grouping as \( (1-1) + (1-1) + \cdots \) gives 0, while \( 1 - (1-1) - (1-1) - \cdots \) gives 1. Which is right? Neither: the series does not converge, and only a precise definition of convergence tells us so.
• Continuous but nowhere differentiable functions. In the 1870s Weierstrass constructed a function that is continuous everywhere yet has no tangent anywhere. Intuition said this was impossible.
• Gaps in the rationals. The rational numbers \( 1, 1.4, 1.41, 1.414, \ldots \) get closer and closer together, but their "limit" \( \sqrt{2} \) is not rational. So "a sequence that settles down must have a limit" is false in \( \mathbb{Q} \). It is true in \( \mathbb{R} \) only because of a special property, the completeness axiom.

Real analysis supplies exact definitions so that arguments like these can be decided once and for all.

The Completeness Axiom

A set \( S \subseteq \mathbb{R} \) is bounded above if some number \( M \) satisfies \( x \leq M \) for all \( x \in S \). The smallest such \( M \) is the supremum, \( \sup S \).

Completeness axiom: every non-empty subset of \( \mathbb{R} \) that is bounded above has a supremum in \( \mathbb{R} \).

This single axiom is what separates \( \mathbb{R} \) from \( \mathbb{Q} \). The set \( \{x \in \mathbb{Q} : x^2 < 2\} \) has no supremum in \( \mathbb{Q} \), but in \( \mathbb{R} \) its supremum is \( \sqrt{2} \). Theorems such as the Intermediate Value Theorem and the Monotone Convergence Theorem all depend on it.

Limits of Sequences: The Epsilon-N Definition

A sequence \( (a_n) \) converges to \( L \) if

$$ \forall \varepsilon > 0, \ \exists N \in \mathbb{N} \text{ such that } n > N \implies |a_n - L| < \varepsilon $$

In words: however small a tolerance \( \varepsilon \) someone chooses, eventually every term of the sequence stays within \( \varepsilon \) of \( L \). Your job in a proof is to produce a suitable \( N \) for any given \( \varepsilon \).

A Standard Proof Strategy

1. Rough work: simplify \( |a_n - L| \) and find a condition on \( n \) that makes it less than \( \varepsilon \).
2. Choose \( N \) from that condition.
3. Write the proof forwards, starting from "Let \( \varepsilon > 0 \)".

Worked Example 1: An Epsilon-N Proof

Prove that

$$ \lim_{n \to \infty} \frac{3n + 1}{n + 2} = 3 $$

Rough work. Simplify the distance from the limit:

$$ \left| \frac{3n + 1}{n + 2} - 3 \right| = \left| \frac{3n + 1 - 3(n + 2)}{n + 2} \right| = \left| \frac{-5}{n + 2} \right| = \frac{5}{n + 2} $$

Since \( n + 2 > n \), we have \( \frac{5}{n+2} < \frac{5}{n} \). So it is enough to make \( \frac{5}{n} < \varepsilon \), that is, \( n > \frac{5}{\varepsilon} \).

Proof. Let \( \varepsilon > 0 \). By the Archimedean property, choose an integer \( N > \frac{5}{\varepsilon} \). Then for all \( n > N \),

$$ \left| \frac{3n+1}{n+2} - 3 \right| = \frac{5}{n+2} < \frac{5}{n} < \frac{5}{N} < \varepsilon $$

Hence the sequence converges to 3. \( \blacksquare \)

For example, if \( \varepsilon = 0.01 \), then \( N = 500 \) works: every term after the 500th is within 0.01 of 3.

Limits of Functions: The Epsilon-Delta Definition

We say \( \lim_{x \to c} f(x) = L \) if

$$ \forall \varepsilon > 0, \ \exists \delta > 0 \text{ such that } 0 < |x - c| < \delta \implies |f(x) - L| < \varepsilon $$

A function is continuous at \( c \) when this holds with \( L = f(c) \) (and the condition \( 0 < |x - c| \) relaxed to \( |x - c| < \delta \)).

Worked Example 2: An Epsilon-Delta Proof

Prove that \( \lim_{x \to 2} x^2 = 4 \).

Rough work. Factorise:

$$ |x^2 - 4| = |x - 2| \, |x + 2| $$

We control \( |x - 2| \) directly, but \( |x + 2| \) varies. So first restrict \( \delta \leq 1 \). Then \( |x - 2| < 1 \) gives \( 1 < x < 3 \), hence \( 3 < x + 2 < 5 \), so \( |x + 2| < 5 \). Now

$$ |x^2 - 4| < 5|x - 2| $$

which is less than \( \varepsilon \) if \( |x - 2| < \frac{\varepsilon}{5} \).

Proof. Let \( \varepsilon > 0 \) and set \( \delta = \min\left(1, \frac{\varepsilon}{5}\right) \). If \( 0 < |x - 2| < \delta \), then \( |x - 2| < 1 \), so \( |x + 2| < 5 \). Also \( |x - 2| < \frac{\varepsilon}{5} \). Therefore

$$ |x^2 - 4| = |x - 2| \, |x + 2| < \frac{\varepsilon}{5} \cdot 5 = \varepsilon $$

Hence \( \lim_{x \to 2} x^2 = 4 \). \( \blacksquare \)

The "minimum" trick is the key idea: one condition tames the awkward factor, the other finishes the job.

Tips for Succeeding in Real Analysis

• Learn definitions word for word. Every proof starts from them, and quantifier order (\( \forall \varepsilon \, \exists N \)) matters.
• Separate rough work from the final proof. Examiners want the clean forward argument.
• Draw pictures. Sketch the \( \varepsilon \)-band around \( L \) to see what you are trying to show.
• Use the triangle inequality, \( |a + b| \leq |a| + |b| \). It appears in almost every proof.
• Test with numbers. Plug in \( \varepsilon = 0.1 \) to check your \( N \) or \( \delta \) really works.
• Do not use what you are proving. Writing "since the limit is 3" inside a proof that the limit is 3 is circular.

Key Takeaways

• Rigor resolves paradoxes and false intuitions that informal calculus cannot.
• The completeness axiom is what makes \( \mathbb{R} \) different from \( \mathbb{Q} \).
• To prove \( a_n \to L \), find an \( N \) for every \( \varepsilon \); to prove \( f(x) \to L \), find a \( \delta \) for every \( \varepsilon \).
• Do rough work backwards, then write the proof forwards.
• Bound awkward factors by first restricting \( \delta \), then take a minimum.

If epsilon-delta proofs or a real analysis assignment are slowing you down, send your questions to Amathify via WhatsApp or the "Ask a Question" form. Our PhD scholars will write clear, rigorous step-by-step solutions for you.