Calculus II: Comparison Tests for Series

Determining whether an infinite series converges or diverges is a fundamental skill in engineering mathematics, with direct applications to signal processing, control theory, and numerical methods. The comparison tests provide a powerful, logical framework for this analysis by leveraging the known behavior of simpler series to deduce the behavior of more complex ones. Mastering these tests is less about memorization and more about developing strategic thinking—a skill crucial for efficiently solving real-world problems where infinite sums model system behavior, error terms, or cumulative effects.

The Logic of Comparison: The Direct Comparison Test

The foundation of this approach is the Direct Comparison Test. This test states that for two series with non-negative terms, $0\le a_n\le b_n$ for all $n$:

1. If the "larger" series $\sum b_n$ converges, then the "smaller" series $\sum a_n$ also converges. Think of it as a finite budget: if you know the total cost of a more expensive set of items is finite, then the total cost of a cheaper set must also be finite.
2. If the "smaller" series $\sum a_n$ diverges, then the "larger" series $\sum b_n$ also diverges. This is the contrapositive: if the cheaper set of items has an infinite total cost, then the more expensive set certainly does.

The critical step is establishing a correct inequality. For example, consider the series ${\sum }_{n=1}^{\infty }\frac{1}{n^2+5}$. We want to show it converges. We know that $\frac{1}{n^2+5}<\frac{1}{n^2}$ for all $n\ge 1$. The series $\sum \frac{1}{n^2}$ is a convergent p-series (where $p=2>1$). Since our series is term-by-term smaller than a known convergent series, the Direct Comparison Test tells us $\sum \frac{1}{n^2+5}$ also converges.

Refining the Comparison: The Limit Comparison Test

Often, finding a strict, correct inequality can be tricky. The terms of our series might oscillate above and below a good comparison series. This is where the more versatile Limit Comparison Test shines. For series with positive terms $a_n$ and $b_n$, you compute the limit: $$L=\underset{n\to \infty }{\lim }\frac{a_n}{b_n}.$$

If $L$ is a finite, positive number ($0<L<\infty$), then the two series $\sum a_n$ and $\sum b_n$ share the same fate—they either both converge or both diverge. This test is exceptionally useful when the terms of your series and the comparison series are essentially proportional for large $n$.

For example, analyze ${\sum }_{n=1}^{\infty }\frac{3n^2-1}{5n^4+2n+7}$. For large $n$, the dominant terms dictate behavior: the numerator behaves like $3n^2$ and the denominator like $5n^4$. Thus, $a_n$ behaves like $(3n^2)/(5n^4)=3/(5n^2)$. We therefore choose the comparison series $b_n=1/n^2$. Now compute the limit: $$L=\underset{n\to \infty }{\lim }\frac{(3n^2-1)/(5n^4+2n+7)}{1/n^2}=\underset{n\to \infty }{\lim }\frac{3n^4-n^2}{5n^4+2n+7}=\frac{3}{5}.$$ Since $L=3/5$ is a finite positive number and $\sum 1/n^2$ converges (p-series, $p=2>1$), our original series also converges by the Limit Comparison Test.

Strategic Selection of a Comparison Series

Choosing the right series for comparison is the core skill. Your go-to toolkit should consist of:

• The p-series, $\sum 1/n^p$: Converges if $p>1$, diverges if $p\le 1$. This is your most common benchmark.
• Geometric series, $\sum a{r}^{n-1}$: Converges if $|r|<1$, diverges if $|r|\ge 1$.
• Known divergent series: The harmonic series $\sum 1/n$ (a p-series with $p=1$) is a classic choice for proving divergence via comparison.

The strategy is to ignore lower-order terms and constant coefficients in your $a_n$ to identify the dominant algebraic growth structure. For $\frac{7n^3}{\sqrt{2n^7-10}}$, focus on $n^3/\sqrt{n^7}=n^3/n^{3.5}=1/n^{0.5}$. This suggests comparison to $b_n=1/n^{0.5}$, a divergent p-series ($p=0.5<1$). You would then use the Limit Comparison Test to confirm.

Developing Skill and Intuition Through Workflow

Efficient analysis follows a mental checklist. First, always check the $n^{th}$ Term Test for Divergence: if ${\lim }_{n\to \infty }{a}_n\ne 0$, the series diverges immediately. If the terms approach zero, proceed to comparison.

Step 1: Simplify and Identify Dominant Behavior. Strip away constants and lower-order terms in the numerator and denominator. What is the resulting power of $n$? For $\frac{4n^2+9n}{n^4+1}$, the core is $n^2/n^4=1/n^2$.

Step 2: Choose a Benchmark p-series. The dominant behavior $1/n^p$ gives you your benchmark. Here, $p=2$, so $\sum 1/n^2$ is the candidate.

Step 3: Select the Appropriate Test.

• Use the Direct Comparison Test if you can easily establish a simple inequality (e.g., making a denominator smaller to make the fraction larger).
• Use the Limit Comparison Test in almost all other algebraic cases, especially when terms are rational functions of $n$. It is more robust and often requires less intricate inequality work.

Step 4: Execute and Conclude. Perform the limit calculation or verify the inequality, then state the conclusion clearly in the context of the benchmark series.

Common Pitfalls

1. Applying the Direct Comparison Test in the Wrong Direction. This is the most frequent error. Remember: if $a_n\le b_n$ and $\sum b_n$ diverges, you cannot conclude anything about $\sum a_n$. A smaller series than a divergent one could still converge or diverge. Similarly, if $a_n\ge b_n$ and $\sum b_n$ converges, you cannot conclude anything.

• Correction: Always double-check the logic. To prove convergence, you must bound your series below a convergent series. To prove divergence, you must bound your series above a divergent series.

1. Incorrectly Applying the Limit Comparison Test with $L=0$ or $L=\infty$. The test is conclusive only if $0<L<\infty$. If $L=0$ and $\sum b_n$ converges, then $\sum a_n$ converges (because $a_n$ is eventually much smaller). However, if $L=0$ and $\sum b_n$ diverges, you learn nothing. The opposite is true for $L=\infty$.

• Correction: If you get $L=0$ or $\infty$, you have likely chosen a comparison series of the wrong order of magnitude. Go back to Step 1 and re-analyze the dominant growth rate.

1. Neglecting to Verify Positivity. Both comparison tests require positive terms (or eventually positive terms). Applying them directly to a series with alternating signs, like $\sum (-1{)}^n{a}_n$, is invalid.

• Correction: For series with both positive and negative terms, you must typically use the Absolute Convergence Test or Alternating Series Test. The comparison tests are applied to the series of absolute values, $\sum |a_n|$, when checking for absolute convergence.

1. Overcomplicating the Comparison Choice. Students often try to match a series too precisely. The power of the Limit Comparison Test is that you only need the asymptotic behavior.

• Correction: Trust the simplification process. For $\frac{5n^3+2}{\sqrt[3]{8n^9-n^2}}$, simplify to $5n^3/\sqrt[3]{8n^9}=(5n^3)/(2n^3)=5/2$. This constant behavior suggests the comparison series $b_n=1$ (or $1/n^0$), which is a divergent p-series ($p=0$).

Summary

• The Direct Comparison Test relies on establishing a term-by-term inequality to compare an unknown series to one with known behavior. Its logic is intuitive but can be algebraically challenging to set up correctly.
• The Limit Comparison Test is often more practical, requiring only that the ratio of terms approaches a finite, positive limit. If it does, the two series share the same convergence outcome.
• Strategic success hinges on simplifying the general term $a_n$ to identify its dominant growth rate, typically expressed as $1/n^p$, and then selecting the corresponding p-series as your benchmark for comparison.
• Developing proficiency requires a disciplined workflow: simplify, choose a benchmark p-series, select the most efficient test (often Limit Comparison), execute carefully, and avoid common logical and algebraic traps.
• These tests are indispensable tools for engineers, transforming the analysis of complex, real-world infinite processes into manageable comparisons with well-understood mathematical models.