AP Calculus BC: Series Convergence Tests and Taylor Series

Understanding infinite series is the gateway from basic calculus to advanced mathematics, physics, and engineering. For AP Calculus BC students, mastering series convergence tests and Taylor series is not just about passing the exam; it’s about acquiring the tools to approximate complex functions, solve otherwise intractable problems, and model real-world phenomena with precision. This unit demands a shift from procedural calculation to strategic reasoning, where choosing the right test is as important as executing it correctly.

The Strategic Toolkit: Series Convergence Tests

An infinite series is the sum of the terms of an infinite sequence, written as ${\sum }_{n=1}^{\infty }{a}_n$. The central question is whether this sum approaches a finite limit (converges) or grows without bound or oscillates (diverges). Your strategy begins with recognizing the form of the series' terms.

Start with the simplest tests. A geometric series ${\sum }_{n=0}^{\infty }a{r}^n$ converges if and only if its common ratio $|r|<1$, and its sum is $\frac{a}{1-r}$. The p-series ${\sum }_{n=1}^{\infty }\frac{1}{n^p}$ converges if $p>1$ and diverges if $p\le 1$. These are your quick checks. For the integral test, you verify that the function $f(x)$ corresponding to $a_n$ is continuous, positive, and decreasing for $x\ge N$. If so, the series $\sum a_n$ and the improper integral ${\int }_N^{\infty }f(x)dx$ either both converge or both diverge. This test is powerful for series involving logarithms, like $\sum \frac{1}{n\ln n}$.

When a series resembles a known convergent or divergent series, use comparison tests. The Direct Comparison Test requires you to find a known series with terms $b_n$ such that $0\le a_n\le b_n$ for all $n$ (for convergence) or $0\le b_n\le a_n$ (for divergence). The Limit Comparison Test is often more flexible: if ${\lim }_{n\to \infty }\frac{a_n}{b_n}=L$, where $0<L<\infty$, then $\sum a_n$ and $\sum b_n$ behave the same. For example, to test $\sum \frac{n+5}{2n^3-1}$, compare it to the convergent p-series $\sum \frac{1}{n^2}$.

For series involving factorials, exponentials, or $n^{th}$ powers, the Ratio Test and Root Test are your go-to tools. The Ratio Test examines ${\lim }_{n\to \infty }|\frac{a_{n+1}}{a_n}|=L$. If $L<1$, convergence; $L>1$, divergence; $L=1$, the test is inconclusive. This test excels for series like $\sum \frac{n!}{10^n}$ or $\sum \frac{x^n}{n!}$. The Root Test uses ${\lim }_{n\to \infty }\sqrt[n]{|a_n|}=L$ with the same rules and is particularly effective when terms are raised to the $n^{th}$ power, like $\sum (\frac{3n}{4n+1}{)}^n$.

Finally, for series where terms alternate in sign, written as ${\sum }_{n=1}^{\infty }(-1{)}^{n-1}{b}_n$ with $b_n>0$, apply the Alternating Series Test. The series converges if: 1) $b_{n+1}\le b_n$ for all $n$ (decreasing), and 2) ${\lim }_{n\to \infty }{b}_n=0$. Crucially, this test establishes conditional convergence. If $\sum |a_n|$ converges, the series is absolutely convergent, a stronger form of convergence.

Representing Functions: Taylor and Maclaurin Series

A Taylor series is a representation of a function as an infinite sum of terms calculated from the function's derivatives at a single point. The Taylor series for $f(x)$ centered at $x=a$ is: $$f(x)=\sum _{n=0}^{\infty }\frac{f^{(n)}(a)}{n!}(x-a{)}^n=f(a)+f^{\prime }(a)(x-a)+\frac{f^{\prime \prime }(a)}{2!}(x-a{)}^2+...$$ When the center is $a=0$, the series is called a Maclaurin series. These series are powerful because they allow us to approximate transcendental functions (like $e^x$, $\sin x$) with polynomials.

Constructing a Taylor polynomial of degree $n$, denoted $P_n(x)$, involves truncating the infinite series after the $(x-a{)}^n$ term. For example, the third-degree Taylor polynomial for $f(x)=\ln (1+x)$ centered at $a=0$ is $P_3(x)=x-\frac{x^2}{2}+\frac{x^3}{3}$. You derive this by computing $f(0)=0,f^{\prime }(0)=1,f^{\prime \prime }(0)=-1,f^{\prime \prime \prime }(0)=2$ and plugging into the formula.

The Lagrange error bound provides a way to quantify the accuracy of this polynomial approximation. If $|f^{(n+1)}(x)|\le M$ for all $x$ between $a$ and your input $x$, then the error $|R_n(x)|=|f(x)-P_n(x)|$ satisfies: $$|R_n(x)|\le \frac{M}{(n+1)!}|x-a{|}^{n+1}$$ For instance, to approximate $e^{0.1}$ using the third-degree Maclaurin polynomial for $e^x$, you know $f^{(4)}(x)=e^x$. On the interval $[0,0.1]$, the maximum of $e^x$ is $e^{0.1}<1.11$. Thus, $|R_3(0.1)|\le \frac{1.11}{4!}(0.1{)}^4\approx 0.000004625$, guaranteeing high accuracy.

You are not expected to derive every series from scratch. Instead, you must skillfully manipulate known Maclaurin series for common functions to derive new representations. Key series to know by heart are: $$e^x=\sum _{n=0}^{\infty }\frac{x^n}{n!}=1+x+\frac{x^2}{2!}+...\text{for all }x$$ $$\sin x=\sum _{n=0}^{\infty }\frac{(-1{)}^n{x}^{2n+1}}{(2n+1)!}=x-\frac{x^3}{3!}+\frac{x^5}{5!}-...\text{for all }x$$ $$\cos x=\sum _{n=0}^{\infty }\frac{(-1{)}^n{x}^{2n}}{(2n)!}=1-\frac{x^2}{2!}+\frac{x^4}{4!}-...\text{for all }x$$ $$\frac{1}{1-x}=\sum _{n=0}^{\infty }{x}^n=1+x+x^2+...\text{for }|x|<1$$

From these, you can find series for $e^{-x^2}$ by substitution, for $x\cos (3x)$ by multiplication, or for $\sin (x^2)$ by replacing $x$ with $x^2$. You can also integrate or differentiate known series term-by-term within their interval of convergence. For example, integrating the geometric series $\frac{1}{1-x}=\sum x^n$ gives the series for $-\ln |1-x|$.

The Radius and Interval of Convergence

For any power series $\sum c_n(x-a{)}^n$, there is a nonnegative number $R$ called the radius of convergence. Within the interval $|x-a|<R$, the series converges absolutely; outside, it diverges. The set of all $x$ for which the series converges is the interval of convergence, which requires checking the endpoints $x=a\pm R$ separately with other tests (like the alternating series test or p-series test).

To find $R$, the Ratio Test is typically most efficient. You set up the limit ${\lim }_{n\to \infty }|\frac{c_{n+1}(x-a{)}^{n+1}}{c_n(x-a{)}^n}|=|x-a|\cdot {\lim }_{n\to \infty }|\frac{c_{n+1}}{c_n}|<1$ and solve for $|x-a|$. For the series $\sum \frac{(x-3{)}^n}{n\cdot 5^n}$, you would find: $$\underset{n\to \infty }{\lim }|\frac{(x-3{)}^{n+1}}{(n+1)5^{n+1}}\cdot \frac{n{5}^n}{(x-3{)}^n}|=|x-3|\cdot \frac{1}{5}\cdot \underset{n\to \infty }{\lim }\frac{n}{n+1}=\frac{|x-3|}{5}$$ Setting $\frac{|x-3|}{5}<1$ gives $|x-3|<5$, so $R=5$. You then test $x=-2$ and $x=8$ separately to determine if the interval is $(-2,8)$, $[-2,8)$, etc.

Common Pitfalls

1. Misapplying the Alternating Series Test: The most common error is checking only that $\lim b_n=0$ without verifying the decreasing condition ($b_{n+1}\le b_n$). For a series like $\sum (-1{)}^n\frac{\ln n}{n}$, the limit is zero, but the terms are not decreasing for very small $n$ (check with calculus). You must ensure the decrease holds eventually, typically for all $n\ge N$.
2. Confusing Sequence vs. Series Behavior: Remember, if ${\lim }_{n\to \infty }{a}_n\ne 0$, the series $\sum a_n$ must diverge (the Divergence Test). However, the converse is false: ${\lim }_{n\to \infty }{a}_n=0$ does not guarantee convergence (e.g., the harmonic series). This is a frequent trap in multiple-choice questions.
3. Incorrect Handling of Series Manipulations: When substituting into a known series, the interval of convergence changes. If you find the series for $f(5x)$ from the series for $f(x)$ with interval $|x|<R$, the new interval is $|5x|<R$, or $|x|<R/5$. Forgetting to adjust the interval is a critical oversight.
4. Lagrange Error Bound Missteps: The common mistake is using the wrong derivative for $M$. You must bound the next derivative, $|f^{(n+1)}(x)|$, on the interval between the center $a$ and the $x$-value you're approximating. Using the $n^{th}$ derivative or bounding over the wrong interval will yield an incorrect error estimate.

Summary

• Convergence testing is a strategic process: start with the Divergence Test (check if terms go to zero), then identify the series form to select the most efficient specific test (geometric, p-series, alternating) or advanced test (comparison, ratio, root, integral).
• A Taylor series centered at $a$ represents a function as an infinite polynomial based on its derivatives at $a$, with the Maclaurin series being the special case where $a=0$. The Taylor polynomial $P_n(x)$ is a finite truncation used for approximation.
• The Lagrange error bound $|R_n(x)|\le \frac{M}{(n+1)!}|x-a{|}^{n+1}$ quantifies the maximum error of a Taylor polynomial approximation, where $M$ is a bound on the $(n+1{)}^{th}$ derivative.
• You must be able to construct new series representations by substituting, differentiating, or integrating known Maclaurin series for $e^x$, $\sin x$, $\cos x$, and $\frac{1}{1-x}$.
• Every power series has a radius of convergence $R$, found most often via the Ratio Test. The interval of convergence is found by testing the endpoints $x=a\pm R$ separately with other convergence tests.
• On the AP exam, clarity of reasoning is paramount. Always state which test you are using and show the logical conditions that lead to your conclusion about convergence or divergence.