When to Use Limit Comparison Test: A 2024 Guide

The limit comparison test is a method in calculus used to determine whether an infinite series converges or diverges by comparing it to another series with known behavior. It is most useful when the terms of two series behave similarly as $ n $ becomes large, especially when direct comparison is difficult. If the limit of the ratio of their terms approaches a finite positive number, both series either converge or diverge together. This test works well for series involving rational functions, roots, or powers, where long-term behavior is predictable.

How the Limit Comparison Test Works

The test applies to series $ \sum a_n $ and $ \sum b_n $, where $ a_n \geq 0 $ and $ b_n > 0 $ for all $ n $. Compute the limit:

$$ L = \lim_{n \to \infty} \frac{a_n}{b_n} $$

If $ L $ is a finite positive number (i.e., $ 0 < L < \infty $), then both series share the same convergence behavior. That means:

• If $ \sum b_n $ converges, so does $ \sum a_n $.
• If $ \sum b_n $ diverges, so does $ \sum a_n $.

This avoids the need to establish term-by-term inequalities, which makes it more flexible than the direct comparison test. The key is choosing a $ b_n $ that captures the dominant behavior of $ a_n $.

When to Use the Limit Comparison Test

Use this test when the general term of your series resembles a simpler one for large $ n $, particularly with algebraic expressions. It works best for:

• Rational functions (e.g., ratios of polynomials)
• Series with roots or powers, such as $ \sqrt{n} $ or $ n^p $
• Terms where numerator and denominator grow at comparable rates

In these cases, comparing to a p-series $ \sum \frac{1}{n^p} $ is often effective. Recall that a p-series converges if $ p > 1 $ and diverges if $ p \leq 1 $.

For example, consider the series:

$$ \sum \frac{3n + 1}{n^3 + 2n} $$

The dominant terms are $ 3n $ in the numerator and $ n^3 $ in the denominator, so the expression behaves like $ \frac{3n}{n^3} = \frac{3}{n^2} $. This suggests using $ b_n = \frac{1}{n^2} $, a convergent p-series with $ p = 2 $.

Now compute the limit:

$$ \lim_{n \to \infty} \frac{(3n + 1)/(n^3 + 2n)}{1/n^2} = \lim_{n \to \infty} \frac{(3n + 1) \cdot n^2}{n^3 + 2n} = \lim_{n \to \infty} \frac{3n^3 + n^2}{n^3 + 2n} $$

Divide numerator and denominator by $ n^3 $:

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

Since the limit is 3 — a finite positive number — and $ \sum \frac{1}{n^2} $ converges, the original series also converges.

When Not to Use the Test

The limit comparison test has limitations. It does not apply if:

• The limit $ \frac{a_n}{b_n} $ is zero or infinity, unless you interpret the result carefully.
• The series has alternating signs. For such cases, use the alternating series test or check for absolute convergence.
• The terms are not eventually positive. The test requires non-negative $ a_n $ and positive $ b_n $.

For example, if $ \lim_{n \to \infty} \frac{a_n}{b_n} = 0 $ and $ \sum b_n $ converges, then $ \sum a_n $ converges. But if $ \sum b_n $ diverges, no conclusion can be drawn. Similarly, if the limit is infinite and $ \sum b_n $ diverges, then $ \sum a_n $ diverges — but again, nothing follows if $ \sum b_n $ converges.

The test also struggles with rapidly growing terms like factorials or exponentials. For instance, series such as $ \sum \frac{n!}{n^n} $ are better handled by the ratio test, since factorial growth doesn’t align well with polynomial-based comparisons.

Choosing an inappropriate $ b_n $ can also lead to inconclusive results. Suppose you try to compare $ \sum \frac{1}{n\sqrt{n}} = \sum \frac{1}{n^{3/2}} $ to $ \sum \frac{1}{n} $. Then:

$$ \lim_{n \to \infty} \frac{1/(n^{3/2})}{1/n} = \lim_{n \to \infty} \frac{1}{\sqrt{n}} = 0 $$

Even though $ \sum \frac{1}{n} $ diverges, the zero limit gives no information. But we know $ \sum \frac{1}{n^{3/2}} $ converges because it’s a p-series with $ p = 1.5 > 1 $. The issue was choosing a comparison series that decays too slowly.

A good $ b_n $ should reflect the asymptotic behavior of $ a_n $. Focus on the highest-degree terms in polynomials or the fastest-growing parts of expressions. For logarithmic terms, note that $ \ln n $ grows slower than any positive power of $ n $, so $ \frac{1}{n \ln n} $ decays slower than $ \frac{1}{n^{1+\varepsilon}} $, but comparing it directly to $ \frac{1}{n} $ may fail because:

$$ \lim_{n \to \infty} \frac{1/(n \ln n)}{1/n} = \lim_{n \to \infty} \frac{1}{\ln n} = 0 $$

Again, the limit is zero, so the test is inconclusive. In such cases, the integral test is often more effective.

Key Takeaways

• Use the limit comparison test when two series have similar long-term behavior and both have positive terms.
• The test requires $ \lim_{n \to \infty} \frac{a_n}{b_n} $ to be a finite positive number. If so, both series converge or diverge together.
• It is most effective for rational functions, roots, and polynomial-like terms, especially when comparing to a p-series.
• Avoid using it for alternating series, factorials, or exponential terms — other tests like the ratio or alternating series test are better suited.
• Always verify the convergence of the comparison series first, typically using p-series or geometric series as benchmarks Paul’s Online Math Notes.
• Choose $ b_n $ based on dominant behavior: keep the highest powers in polynomials, simplify complex expressions, and ensure growth rates are comparable.
• A limit of zero or infinity may give partial or no information, so interpret carefully and consider alternative tests if needed.