Mathematical analysis Study Guide

📖 Core Concepts Mathematical Analysis – Study of continuous objects (functions, limits, series) and the rigorous foundations of calculus. Metric Space \((X,d)\) – Set \(X\) with a distance function \(d\) satisfying non‑negativity, identity of indiscernibles, symmetry, and triangle inequality. Sequence & Limit – A sequence \((an)\) is a function \(n\mapsto an\). It converges to \(L\) if \(\lim{n\to\infty}d(an,L)=0\). Continuity (topological definition) – For every neighbourhood \(V\) of \(f(c)\) there is a neighbourhood \(U\) of \(c\) with \(f(U)\subseteq V\). Real vs. Complex vs. Functional Analysis – Real: properties of \(\mathbb R\)‑valued functions; Complex: holomorphic functions on \(\mathbb C\); Functional: vector spaces with limits (norms, inner products) and linear operators. Measure Theory – Assigns a non‑negative size \(\mu\) to sets, with \(\mu(\varnothing)=0\) and countable additivity on disjoint families. --- 📌 Must Remember Metric axioms: \(d(a,b)\ge0\); \(d(a,b)=0\iff a=b\); \(d(a,b)=d(b,a)\); \(d(a,c)\le d(a,b)+d(b,c)\). Convergence criterion: \(\forall\varepsilon>0,\ \exists N\) such that \(n\ge N\Rightarrow d(an,L)<\varepsilon\). Continuity \(\varepsilon\)–\(\delta\) (in metric spaces): \(\forall\varepsilon>0,\ \exists\delta>0\) s.t. \(d(x,c)<\delta\Rightarrow d(f(x),f(c))<\varepsilon\). Cauchy sequence: \(\forall\varepsilon>0,\ \exists N\) s.t. \(m,n\ge N\Rightarrow d(am,an)<\varepsilon\). (In complete metric spaces, every Cauchy sequence converges.) Riemann vs. Lebesgue integration – Lebesgue integrates more functions by measuring “size” of level sets; it dominates Riemann. Banach space – Complete normed vector space (central in functional analysis). Hilbert space – Complete inner‑product space (generalizes Euclidean space). --- 🔄 Key Processes Proving a limit in a metric space Choose \(\varepsilon>0\). Find a \(\delta\) (or an \(N\) for sequences) using the definition of the function/sequence. Show that the distance condition holds. Showing continuity at a point Start with an arbitrary \(\varepsilon>0\). Derive a \(\delta\) (or neighbourhood \(U\)) that forces the image to stay inside the \(\varepsilon\)-ball around \(f(c)\). Verifying a metric Check the four axioms for the proposed distance function. Establishing completeness Take an arbitrary Cauchy sequence. Show it converges to a limit inside the space (often by constructing the limit or invoking known theorems). --- 🔍 Key Comparisons Real Analysis vs. Complex Analysis Real: Limits, continuity, differentiability defined on \(\mathbb R\). Complex: Uses holomorphic (analytic) functions; Cauchy–Riemann equations give stronger differentiability. Riemann Integration vs. Lebesgue Integration Riemann: Partitions domain, sums heights × widths; fails for many “pathological” sets. Lebesgue: Partitions codomain, measures pre‑images; handles more functions, especially with infinite discontinuities. Metric Space vs. Topological Space Metric: Distance function provides a concrete way to talk about “closeness”. Topological: Only open sets are specified; every metric space induces a topology, but not every topology comes from a metric. Banach Space vs. Hilbert Space Banach: Complete normed space (norm may not arise from an inner product). Hilbert: Norm comes from an inner product; enjoys orthogonal projection theory. --- ⚠️ Common Misunderstandings “Every continuous function is uniformly continuous.” – True on compact domains only; not on all metric spaces. “If a sequence is bounded, it converges.” – Only true in ℝ when the sequence is monotone (Monotone Convergence Theorem). “All metrics satisfy the triangle inequality with equality only when points are collinear.” – Equality can occur in many metric spaces (e.g., discrete metric). “Lebesgue measure equals length for every set.” – Holds for measurable sets; non‑measurable sets have no defined Lebesgue measure. --- 🧠 Mental Models / Intuition Metric space – Think of a “ruler” that works for any two points, even in abstract settings. Cauchy sequence – The terms get arbitrarily close to each other, not necessarily to a known point; completeness guarantees a hidden limit. Continuity – “No sudden jumps”: small input wiggle → small output wiggle. Visualize stretching a rubber sheet without tearing. Lebesgue integration – Paint the graph: measure the area by stacking horizontal slices (level sets) rather than vertical rectangles. --- 🚩 Exceptions & Edge Cases Discrete metric (\(d(x,y)=1\) for \(x\neq y\)) – Every subset is open; every sequence converges only if eventually constant. Non‑complete spaces – E.g., \((\mathbb Q,|\cdot|)\) has Cauchy sequences that converge to irrational limits outside the space. Uniform continuity fails on unbounded intervals – \(f(x)=x^2\) is continuous on \(\mathbb R\) but not uniformly continuous. --- 📍 When to Use Which Choose Lebesgue integration when the integrand has many discontinuities or when you need limit‑exchange theorems (Dominated Convergence, Fubini). Apply Banach Fixed‑Point Theorem in a complete metric space with a contraction map to guarantee a unique solution to an equation. Use Cauchy‑Schwarz inequality in inner‑product (Hilbert) spaces to bound inner products and prove convergence of series. Select Fourier analysis (harmonic analysis) when a problem involves periodicity or signal decomposition. --- 👀 Patterns to Recognize “ε‑δ” language → limit or continuity proof. “Monotone + bounded → convergent” → often appears in real analysis sequences. “Complete + Cauchy → limit exists” – hallmark of Banach/Hilbert space arguments. “Linear + bounded operator” → automatically continuous in normed spaces. --- 🗂️ Exam Traps Distractor: “Every bounded sequence has a convergent subsequence.” – True only in ℝⁿ (Bolzano‑Weierstrass), not in arbitrary metric spaces. Trap: Confusing “uniform continuity” with “continuity”. Uniform continuity is stronger; look for domain compactness. Misleading answer: “Lebesgue integral always equals Riemann integral for the same function.” – Only when the function is Riemann‑integrable (i.e., its set of discontinuities has measure zero). False equivalence: “Metric ⇒ normed space.” – Norm induces a metric, but a metric need not come from any norm. ---