AP Calculus BC: Taylor Series Representations

In calculus, we often face functions that are difficult to integrate, differentiate, or evaluate directly. What if you could replace a complex function with an infinite polynomial that behaves identically, at least locally? This is the power of the Taylor series, a cornerstone of AP Calculus BC that transforms analysis and enables approximations essential to engineering, physics, and computer science. Mastering Taylor series allows you to represent transcendental functions like $e^x$ and $\sin (x)$ with simple arithmetic operations, solve otherwise intractable differential equations, and understand the deep connection between a function's local derivatives and its global behavior.

The Foundation: Defining a Taylor Series

A Taylor series is an infinite sum of terms that provides a power series representation of a function centered at a specific point. Formally, the Taylor series for a function $f(x)$ centered at $x=a$ is given by:

$$f(x)=\sum _{n=0}^{\infty }\frac{f^{(n)}(a)}{n!}(x-a{)}^n$$

Here, $f^{(n)}(a)$ denotes the $n$th derivative of $f$ evaluated at the center $x=a$, and $n!$ is the factorial of $n$. The zeroth derivative is the function itself: $f^{(0)}(a)=f(a)$. The expression $(x-a{)}^n$ is the power term that shifts the series' focus to the point $a$. When the center is $a=0$, the series has a special name: it is called a Maclaurin series.

The core idea is that if a function is infinitely differentiable at a point, we can construct a polynomial whose derivatives at that point match the function's derivatives exactly. The more terms we take, the closer this polynomial approximates the original function near the center. This is not just a symbolic trick; for many important functions, the infinite series converges to the function's value for a range of $x$ values.

Deriving Series for Common Functions

The process of constructing a Taylor series involves a clear, step-by-step pattern. Let's derive the Maclaurin series (center $a=0$) for $f(x)=e^x$.

1. Find the derivatives: $f(x)=e^x$, $f^{\prime }(x)=e^x$, $f^{\prime \prime }(x)=e^x$, and in general, $f^{(n)}(x)=e^x$.
2. Evaluate at the center ($a=0$): $f(0)=1$, $f^{\prime }(0)=1$, $f^{\prime \prime }(0)=1$, and $f^{(n)}(0)=1$ for all $n$.
3. Construct the coefficients: The coefficient for the $n$th term is $\frac{f^{(n)}(0)}{n!}=\frac{1}{n!}$.
4. Write the series: $$e^x=\sum _{n=0}^{\infty }\frac{x^n}{n!}=1+x+\frac{x^2}{2!}+\frac{x^3}{3!}+\frac{x^4}{4!}+\cdots$$

This pattern allows you to commit key series to memory. Two other essential Maclaurin series involve trigonometric functions. For $\sin (x)$, the derivatives cycle through $\cos (x)$, $-\sin (x)$, $-\cos (x)$, $\sin (x)$, causing every other term (the even-powered terms) to vanish. The resulting series contains only odd powers with alternating signs:

$$\sin (x)=\sum _{n=0}^{\infty }\frac{(-1{)}^n{x}^{2n+1}}{(2n+1)!}=x-\frac{x^3}{3!}+\frac{x^5}{5!}-\frac{x^7}{7!}+\cdots$$

Similarly, the series for $\cos (x)$ contains only even powers:

$$\cos (x)=\sum _{n=0}^{\infty }\frac{(-1{)}^n{x}^{2n}}{(2n)!}=1-\frac{x^2}{2!}+\frac{x^4}{4!}-\frac{x^6}{6!}+\cdots$$

Another crucial series is for $\ln (1+x)$, which has a limited radius of convergence ($|x|<1$):

$$\ln (1+x)=\sum _{n=1}^{\infty }\frac{(-1{)}^{n+1}{x}^n}{n}=x-\frac{x^2}{2}+\frac{x^3}{3}-\frac{x^4}{4}+\cdots$$

Convergence and the Remainder: When Does the Series Equal the Function?

A critical question is: when does the infinite Taylor series actually converge to the function's value? The interval of convergence is the set of $x$-values for which the series sum equals $f(x)$. The distance from the center $a$ to the boundary of this interval is the radius of convergence, denoted $R$. You typically find $R$ using the Ratio Test on the series terms.

More subtly, even within the radius of convergence, we use a finite number of terms (a Taylor polynomial, $P_n(x)$) for approximation. The error of this approximation is called the remainder, $R_n(x)=f(x)-P_n(x)$. The Lagrange form of the remainder provides a way to bound this error:

$$R_n(x)=\frac{f^{(n+1)}(z)}{(n+1)!}(x-a{)}^{n+1}$$

Here, $z$ is some number between $x$ and $a$. This formula is powerful because it allows you to determine, for example, how many terms of the series for $\sin (0.1)$ you need to guarantee an error less than $10^{-10}$, a common task in engineering and computational applications. Understanding the remainder separates simply writing a series from rigorously using it for approximation.

Applications and Manipulation of Series

Taylor series are not static formulas; they are tools for solving complex problems. A primary application is approximating definite integrals of functions with no elementary antiderivative. For instance, the integral $\int e^{-x^2}dx$ is fundamental in statistics but has no closed form. Using the series for $e^x$ and substituting $-x^2$ for $x$, we get $e^{-x^2}=1-x^2+\frac{x^4}{2!}-\frac{x^6}{3!}+\cdots$. Integrating this series term-by-term yields a polynomial approximation for the integral that is highly accurate within its radius of convergence.

You can also manipulate known series to find new ones through operations like substitution, differentiation, and integration. For example, to find the series for $x\cos (3x^2)$, you would take the known series for $\cos (u)$, substitute $u=3x^2$, and then multiply the entire result by $x$. This technique is far more efficient than computing dozens of derivatives using the product and chain rules. On the AP exam, you are often asked to construct the first few non-zero terms of a Taylor series for a composite function using these manipulation methods, as it tests your conceptual understanding over brute-force calculation.

Common Pitfalls

1. Confusing Radius and Interval of Convergence: Students often find the radius $R$ correctly but then misstate the interval. You must always test the endpoints $x=a\pm R$ separately in the original series, as convergence at endpoints can be conditional or absolute. For the series of $\ln (1+x)$, the radius is $R=1$. Testing $x=1$ gives the convergent alternating harmonic series, but testing $x=-1$ gives the divergent negative harmonic series. Thus, the interval of convergence is $(-1,1]$.

1. Forgetting the Domain of the Original Function: The series for $\ln (1+x)$ converges to $\ln (1+x)$ only for $|x|<1$, with careful inclusion of one endpoint. Writing the equality without stating this condition is a critical error. Always pair the series expression with its condition for convergence.

1. Incorrect Derivative Evaluation During Construction: When building a series from scratch, a single miscalculated derivative propagates through all subsequent terms. Organize your work in a table: list $n$, $f^{(n)}(x)$, $f^{(n)}(a)$, and the coefficient $\frac{f^{(n)}(a)}{n!}$. For centers other than zero ($a\ne 0$), remember the power term is $(x-a{)}^n$, not $x^n$.

1. Misapplying the Lagrange Remainder: The value $z$ in the formula $R_n(x)=\frac{f^{(n+1)}(z)}{(n+1)!}(x-a{)}^{n+1}$ is unknown; it is "some $z$ between $x$ and $a$." To find an error bound, you must find the maximum possible value of $|f^{(n+1)}(z)|$ on that interval. Using the derivative at the endpoint instead of its maximum can lead to an incorrect, non-conservative error estimate.

Summary

• A Taylor series represents a function as an infinite polynomial: $f(x)={\sum }_{n=0}^{\infty }\frac{f^{(n)}(a)}{n!}(x-a{)}^n$. Its coefficients are determined solely by the derivatives of the function at the center $a$.
• You must derive and memorize key Maclaurin series ($a=0$) for $e^x$, $\sin (x)$, $\cos (x)$, and $\ln (1+x)$, as they form the building blocks for constructing more complex series through substitution, differentiation, and integration.
• The radius of convergence $R$ dictates where the series converges to the function. Always test endpoints separately to determine the full interval of convergence.
• The Lagrange remainder, $R_n(x)=\frac{f^{(n+1)}(z)}{(n+1)!}(x-a{)}^{n+1}$, provides a crucial method for bounding the error when using a finite Taylor polynomial to approximate a function.
• Taylor series are indispensable tools for approximating function values, evaluating definite integrals of non-elementary functions, and solving differential equations that model real-world systems in engineering and physics.