Sequences and Limits (Rigorous)

Why This Matters

In calculus, you learned that a sequence converges if its terms "get closer and closer" to some value. But what does "closer and closer" really mean? Real analysis replaces that vague intuition with a precise definition: for every epsilon greater than 0, there exists an N such that for all n greater than or equal to N, the distance from a_n to L is less than epsilon.

A Cauchy sequence is one where the terms get arbitrarily close to each other, even when you do not know the limit in advance. The completeness of R guarantees that every Cauchy sequence converges — this is the key property that makes rigorous convergence theory work. Understanding these definitions is essential for proving theorems about series, continuity, and integration.

Define Terms

Visual Model

Code Example

// Epsilon-N definition in action
// Sequence a_n = 1/n converges to 0
function findN(epsilon) {
  // We need 1/n < epsilon, so n > 1/epsilon
  return Math.ceil(1 / epsilon) + 1;
}

console.log("a_n = 1/n converges to 0:");
for (const eps of [0.1, 0.01, 0.001, 0.0001]) {
  const N = findN(eps);
  console.log(`  eps=${eps}: N=${N}, a_N = ${(1/N).toExponential(3)} < ${eps}`);
}

// Verify convergence: check |a_n - L| < epsilon for all n >= N
function verifyConvergence(seq, L, epsilon, startN, count) {
  for (let n = startN; n < startN + count; n++) {
    const diff = Math.abs(seq(n) - L);
    if (diff >= epsilon) return false;
  }
  return true;
}

const aN = n => 1 / n;
console.log("Verified for eps=0.01:", verifyConvergence(aN, 0, 0.01, 101, 1000));

// Cauchy sequence check
// a_n = sum of 1/k^2 for k=1..n (converges to pi^2/6)
function partialSum(n) {
  let s = 0;
  for (let k = 1; k <= n; k++) s += 1 / (k * k);
  return s;
}

console.log("\nCauchy check for sum 1/k^2:");
for (const eps of [0.1, 0.01, 0.001]) {
  let N = 1;
  while (Math.abs(partialSum(N + 10) - partialSum(N)) >= eps) N++;
  console.log(`  eps=${eps}: terms are within eps after N=${N}`);
}
console.log("Limit (pi^2/6):", (Math.PI * Math.PI / 6).toFixed(8));
console.log("Partial sum n=1000:", partialSum(1000).toFixed(8));

// Divergent sequence: a_n = (-1)^n does not converge
console.log("\na_n = (-1)^n: first 10 terms:");
for (let n = 1; n <= 10; n++) {
  console.log(`  a_${n} = ${Math.pow(-1, n)}`);
}

Interactive Experiment

Try these exercises:

• For a_n = 1/n^2, find the smallest N such that |a_n| is below 0.0001 for all n at least N. Does the N you find match ceil(1/sqrt(epsilon))?
• Compute the first 50 partial sums of the harmonic series 1 + 1/2 + 1/3 + ... Does this sequence appear Cauchy? It is not — it diverges, just very slowly.
• Take a_n = (1 + 1/n)^n. Verify numerically that this is a bounded, increasing sequence. What does it converge to?
• Check whether a_n = sin(n) is Cauchy. Compute |a_100 - a_101|, |a_1000 - a_1001|. Does the difference shrink?
• Construct a sequence in the rationals that is Cauchy but whose limit is irrational (like the decimal expansion of sqrt(2) truncated at n digits).

Quick Quiz

Coding Challenge

Real-World Usage

Rigorous convergence theory appears throughout applied mathematics and engineering:

• Numerical methods: Iterative solvers (Jacobi, Gauss-Seidel, conjugate gradient) must be shown to produce Cauchy sequences to guarantee they converge to the true solution.
• Machine learning: Gradient descent convergence proofs rely on showing that the parameter sequence is Cauchy under appropriate learning rate conditions.
• Signal processing: Fourier series convergence (pointwise, uniform, L^2) all use epsilon-N style arguments from real analysis.
• Control theory: Stability analysis verifies that system states converge, using the same formal framework as sequence limits.
• Financial mathematics: Stochastic processes in quantitative finance use convergence in probability and almost sure convergence, both built on the rigorous limit definition.

Connections