Comparison Tests for Convergence

SECTION 1

Historical Context & Motivation

The question of whether an infinite sum produces a finite value or grows without bound has captivated mathematicians for centuries. Early work by Nicole Oresme in the fourteenth century demonstrated that the harmonic series diverges, yet many other series with similarly shrinking terms do converge. This tension—between series that look similar but behave differently—motivated the development of systematic tools for testing convergence. The comparison tests emerged as one of the earliest and most intuitive of these tools, built on a simple but powerful idea: if you can bound one series by another whose convergence is already established, you can determine the behavior of the unknown series without computing its sum directly.

1350

Oresme's Divergence Proof

Nicole Oresme proved the harmonic series ∑1/n diverges by grouping terms, establishing the first rigorous divergence result and revealing that shrinking terms alone do not guarantee convergence.

1689

Jakob Bernoulli's p-Series

Jakob Bernoulli investigated series of the form ∑1/nᵖ and showed that such series converge for p > 1, providing a rich family of benchmark series against which other series could be compared.

1821

Cauchy's Cours d'analyse

Augustin-Louis Cauchy formalized the Direct Comparison Test and several other convergence criteria in his landmark textbook, placing the study of infinite series on rigorous analytic foundations.

1850s

Limit Comparison Emerges

Mathematicians including Karl Weierstrass refined comparison techniques into the Limit Comparison Test, enabling analysts to draw convergence conclusions even when a direct inequality is difficult to establish.

The central challenge these tests address is this: given a series ∑aₙ with complicated terms, how can we determine its convergence or divergence without finding a closed-form expression for its partial sums? The comparison tests answer this by leveraging the known behavior of simpler series—particularly geometric series and p-series—to draw conclusions about more complex ones. Mastering these tests is essential for the AP Calculus BC exam, where choosing the right convergence strategy often determines whether a problem is tractable.

SECTION 2

Core Principles & Definitions

Both comparison tests rest on a foundational idea from real analysis: the behavior of a series with nonnegative terms is completely determined by whether its partial sums are bounded above. If a nonnegative series has partial sums that remain below some finite ceiling, the series must converge; if its partial sums can be pushed arbitrarily high by those of a divergent series, it must diverge. The two comparison tests formalize this intuition in different ways—one through direct inequalities, the other through asymptotic ratios.

Direct Comparison Test (DCT)

If 0 ≤ aₙ ≤ bₙ for all n ≥ N, then: (1) if ∑bₙ converges, so does ∑aₙ; (2) if ∑aₙ diverges, so does ∑bₙ. The inequality must hold for all sufficiently large n.

Limit Comparison Test (LCT)

If aₙ > 0, bₙ > 0, and lim(n→∞) aₙ/bₙ = L where 0 < L < ∞, then ∑aₙ and ∑bₙ either both converge or both diverge. No direct inequality is needed—only the limiting ratio.

Benchmark Series

The p-series ∑1/nᵖ (converges for p > 1, diverges for p ≤ 1) and the geometric series ∑arⁿ (converges for |r| < 1) serve as the primary comparison partners for most problems.

Nonnegative Terms Required

Both tests require all terms to be nonnegative (or at least eventually nonnegative). For series with mixed signs, apply these tests to ∑|aₙ| to check absolute convergence.

✦ KEY TAKEAWAY

Think of comparison tests like estimating whether a bucket will overflow. If you know a larger stream of water (the comparison series) fills the bucket to a finite level, then a smaller stream (your series) certainly will too. Conversely, if even a smaller stream causes overflow (divergence), a larger stream must also overflow. The Direct Comparison Test requires you to verify the stream sizes directly; the Limit Comparison Test lets you check whether two streams flow at roughly the same rate in the long run.

SECTION 3

Visual Explanation

The following diagram illustrates the logic of the Direct Comparison Test. Two series with nonnegative terms are plotted as bar charts. When the terms of ∑aₙ are bounded above by the terms of a convergent series ∑bₙ, the partial sums of ∑aₙ are trapped below a finite ceiling and therefore converge. The visual makes it clear that the inequality between individual terms translates into an inequality between partial sums.

Each pair of bars at index n shows bₙ (violet, taller) and aₙ (cyan, shorter). Since 0 ≤ aₙ ≤ bₙ and ∑bₙ converges, the cumulative area under the cyan bars is trapped below a finite value, so ∑aₙ also converges.

The pink dashed curve represents the envelope of the bₙ terms. Every cyan bar (aₙ) lies beneath its corresponding violet bar (bₙ), and every violet bar lies beneath the envelope. Because the total area under the violet bars is finite (∑bₙ converges), the total area under the smaller cyan bars must also be finite. This geometric reasoning is precisely what the Direct Comparison Test formalizes.

SECTION 4

Mathematical Framework

Let us state each test precisely and examine the conditions that must be verified before applying them. Both tests assume that we are working with series of nonnegative terms, or at least terms that are eventually nonnegative (i.e., nonnegative for all n beyond some threshold N).

Direct Comparison Test (DCT)

DCT — CONVERGENCE

If 0 ≤ aₙ ≤ bₙ for all n ≥ N and ∑bₙ converges, then ∑aₙ converges.

aₙ = terms of the series under investigation; bₙ = terms of a known convergent series; N = index beyond which the inequality holds.

DCT — DIVERGENCE

If 0 ≤ bₙ ≤ aₙ for all n ≥ N and ∑bₙ diverges, then ∑aₙ diverges.

Here the comparison is flipped: the known divergent series bₙ is smaller, which forces aₙ to also diverge.

Limit Comparison Test (LCT)

LIMIT COMPARISON TEST

If aₙ > 0, bₙ > 0, and lim(n→∞) aₙ / bₙ = L with 0 < L < ∞, then ∑aₙ and ∑bₙ either both converge or both diverge.

L = the limiting ratio; aₙ, bₙ must be strictly positive for all sufficiently large n. If L = 0 or L = ∞, the test still gives partial information (see Section 7).

💡 Why Does the LCT Work?

If aₙ/bₙ → L with 0 < L < ∞, then for large n we have approximately (L/2)bₙ ≤ aₙ ≤ (2L)bₙ. This means the two series are essentially scalar multiples of each other in the tail, so one converges if and only if the other does. The LCT is thus a Direct Comparison Test in disguise, applied to the tail of the series.

SECTION 5

Choosing a Comparison Series

The success of either comparison test depends entirely on selecting an appropriate benchmark series. The general strategy is to examine the dominant behavior of aₙ for large n: strip away lower-order additive terms, logarithmic factors, and constants to identify a simpler expression that captures how fast the terms shrink. This simplified expression becomes your candidate bₙ. The diagram below illustrates a decision flowchart for choosing between the Direct Comparison Test and the Limit Comparison Test based on the structure of the series.

Start by identifying a benchmark series bₙ that matches the dominant behavior of aₙ. If a direct inequality is easy to establish, use the Direct Comparison Test. If the algebra of the inequality is cumbersome—especially when denominators involve sums or differences of polynomial terms—the Limit Comparison Test is usually more efficient.

Common Benchmark Series

Common benchmark series and when to deploy them

Benchmark Series	Converges When	Typical Use
∑ 1/nᵖ (p-series)	p > 1	Rational functions of n, polynomial-dominated terms
∑ arⁿ (geometric)	\|r\| < 1	Exponential decay, series involving rⁿ factors
∑ 1/n (harmonic)	Never (diverges)	Lower bound for showing divergence
∑ 1/(n ln n)	Never (diverges)	Series with logarithmic factors near the convergence boundary

SECTION 6

Worked Examples

Example 1: Direct Comparison Test

Determine whether ∑(n=1 to ∞) 1/(n³ + 5) converges or diverges.

Step 1 — Identify Dominant BehaviorFor large n, the "+5" in the denominator becomes negligible compared to n³. So 1/(n³ + 5) behaves like 1/n³, which is a p-series with p = 3 > 1. We expect convergence.

Step 2 — Establish the InequalitySince n³ + 5 > n³ for all n ≥ 1, we can flip the inequality to obtain 1/(n³ + 5) < 1/n³. Thus 0 < aₙ < bₙ where aₙ = 1/(n³ + 5) and bₙ = 1/n³.

0 < 1/(n³ + 5) < 1/n³ for all n ≥ 1

Step 3 — Apply the DCTThe series ∑1/n³ is a convergent p-series (p = 3 > 1). Since 0 < aₙ < bₙ and ∑bₙ converges, the Direct Comparison Test guarantees that ∑1/(n³ + 5) converges.

∑ 1/(n³ + 5) converges by the Direct Comparison Test.

Example 2: Limit Comparison Test

Determine whether ∑(n=1 to ∞) (3n² + 2)/(n⁴ − n + 7) converges or diverges.

Step 1 — Identify Dominant BehaviorThe dominant term in the numerator is 3n² and in the denominator is n⁴. For large n, aₙ ≈ 3n²/n⁴ = 3/n². So we choose bₙ = 1/n² as our comparison series. Since p = 2 > 1, ∑1/n² converges.

Step 2 — Compute the LimitWe compute lim(n→∞) aₙ/bₙ = lim(n→∞) [(3n² + 2)/(n⁴ − n + 7)] ÷ [1/n²] = lim(n→∞) (3n² + 2) · n² / (n⁴ − n + 7) = lim(n→∞) (3n⁴ + 2n²) / (n⁴ − n + 7). Dividing numerator and denominator by n⁴ yields lim(n→∞) (3 + 2/n²) / (1 − 1/n³ + 7/n⁴) = 3/1 = 3.

L = 3, and 0 < 3 < ∞

Step 3 — Apply the LCTSince L = 3 is a finite, positive number, the Limit Comparison Test tells us that ∑aₙ and ∑bₙ = ∑1/n² share the same convergence behavior. Since ∑1/n² converges (p-series, p = 2 > 1), our original series also converges.

∑ (3n² + 2)/(n⁴ − n + 7) converges by the Limit Comparison Test.

SECTION 7

Strengths, Limitations & Common Pitfalls

Head-to-head comparison of the two comparison tests

Feature	Direct Comparison Test	Limit Comparison Test
Ease of setup	Requires a provable inequality aₙ ≤ bₙ or aₙ ≥ bₙ, which can be tricky	Only requires computing a limit—no inequality needed
Direction sensitivity	The inequality must go the right way: smaller than convergent, or larger than divergent	Direction doesn't matter—any finite positive limit L suffices
Edge cases	Always conclusive when the inequality and known series align	If L = 0 or L = ∞, the test gives only partial information (one direction)
Best suited for	Series where adding/removing terms in the denominator makes the inequality obvious	Rational functions of n, series where dominant-term analysis is natural
Common pitfall	Showing aₙ ≤ bₙ when ∑bₙ diverges proves nothing	Choosing bₙ whose convergence is unknown defeats the purpose

⚠️ Common AP Exam Mistake

Students frequently establish the correct inequality but compare in the wrong direction. Remember: being smaller than a divergent series tells you nothing—a series could be small and still converge. Similarly, being larger than a convergent series tells you nothing—you need the inequality to trap the unknown series beneath a convergent ceiling (or above a divergent floor).

✦ KEY TAKEAWAY

When you're stuck deciding between DCT and LCT, try the LCT first. The limit computation is often more mechanical than establishing an inequality, and on the AP exam, the LCT tends to be the more reliable workhorse for rational expressions. Reserve the DCT for cases where the inequality is obvious—for instance, when you can simply drop a positive term from the denominator to make the fraction larger.

SECTION 8

Connections to Other Convergence Tests

The comparison tests sit within a broader ecosystem of convergence tests, each suited to different structural features of a series. Understanding how these tests relate to one another helps you choose the most efficient approach on the AP exam. The table below positions the comparison tests alongside other key tests in the BC curriculum.

Comparison tests in the context of the AP Calculus BC convergence toolkit

Test	Best For	Relationship to Comparison Tests
Ratio Test	Factorials, exponentials, products	Implicitly compares to a geometric series via the limiting ratio
Root Test	n-th power expressions	Also an implicit geometric comparison; often equivalent to the Ratio Test
Integral Test	Monotone decreasing functions of n	Compares the series to an improper integral; can prove p-series convergence that benchmarks the comparison tests
Alternating Series Test	Series with alternating signs	Handles conditional convergence, which comparison tests cannot address directly
Comparison / Limit Comparison	Rational, algebraic expressions of n	The most direct approach for series whose terms resemble p-series or geometric series

In more advanced analysis courses, comparison arguments generalize naturally into notions such as asymptotic equivalence and big-O / little-o notation. The Limit Comparison Test is essentially the statement that if aₙ ~ Lbₙ (asymptotic equivalence), the two series share convergence behavior. These ideas appear throughout real analysis, probability theory, and even computer science in the analysis of algorithms, making the comparison tests a gateway to far-reaching mathematical tools.

SECTION 9

Practice Problems

Suppose aₙ > 0, bₙ > 0, and aₙ ≤ bₙ for all n ≥ 1. Which of the following conclusions is valid?

Determine whether the series ∑(n=1 to ∞) 1/(2ⁿ + n) converges or diverges.

Using the Limit Comparison Test with an appropriate p-series, determine the convergence or divergence of ∑(n=1 to ∞) (n + 4)/(n³ − 2n + 1).

PROBLEM 4 — APPLIED

Consider the series ∑(n=2 to ∞) 1/(n² − 1). (a) Use partial fraction decomposition to write 1/(n² − 1) in a simpler form. (b) Use the Limit Comparison Test with an appropriate p-series to determine whether the series converges or diverges. Justify your answer completely. (c) The series is also telescoping. Find the exact sum S of the series. (d) Explain why the Direct Comparison Test with bₙ = 1/n² would also establish convergence, and set up the required inequality.

PROBLEM 5 — CRITICAL THINKING

Let aₙ = 1/(n ln n) for n ≥ 2. (a) A student attempts to use the Limit Comparison Test with bₙ = 1/n to determine the convergence of ∑aₙ. Compute lim(n→∞) aₙ/bₙ and explain what the result tells us (or fails to tell us). (b) Explain why neither the DCT with bₙ = 1/n nor the DCT with bₙ = 1/n² is sufficient to determine the convergence or divergence of ∑1/(n ln n). (c) Name a test that can determine the convergence or divergence of ∑1/(n ln n), state the conclusion, and give a one-sentence justification.

SUMMARY

Summary & Key Concepts

The Direct Comparison Test and the Limit Comparison Test are powerful tools for determining the convergence or divergence of series with nonnegative terms. The DCT requires a provable inequality between the given series and a known benchmark series: if 0 ≤ aₙ ≤ bₙ and ∑bₙ converges, then ∑aₙ converges; if 0 ≤ bₙ ≤ aₙ and ∑bₙ diverges, then ∑aₙ diverges. The LCT requires only that lim(n→∞) aₙ/bₙ = L where 0 < L < ∞; the two series then share the same convergence behavior.

Success with these tests hinges on choosing the right comparison series, typically a p-series or geometric series, by analyzing the dominant behavior of the general term for large n. Watch out for the wrong-direction pitfall in the DCT: being smaller than a divergent series or larger than a convergent series yields no conclusion. When establishing a direct inequality proves difficult, the LCT is almost always the more efficient path.