Alternating Series Error Bound
students, imagine you are watching a bouncing ball that keeps losing height every time it hits the ground 🏀. The drops get smaller and smaller, and at some point you want a quick way to estimate how high it will bounce next without adding an endless list of values. In AP Calculus BC, that kind of “good enough” estimate is exactly where the Alternating Series Error Bound helps. It tells you how far your partial sum could be from the true value when you stop after a few terms.
Objectives:
- Explain what an alternating series is and why its error bound works.
- Use the Alternating Series Error Bound to estimate accuracy.
- Connect the idea to convergence, series approximations, and error analysis.
- Recognize when this tool is useful on AP Calculus BC problems.
What Makes a Series Alternating?
An alternating series is a series whose terms switch sign back and forth. A common form is
$$\sum_{n=1}^{\infty} (-1)^{n-1}a_n$$
where each $a_n$ is positive. The signs alternate like $+$, $-$, $+$, $-$, and so on. For example,
$$1-\frac{1}{2}+\frac{1}{3}-\frac{1}{4}+\cdots$$
is alternating.
The Alternating Series Test says that such a series converges if:
- $a_n$ decreases eventually, and
- $\lim_{n\to\infty} a_n = 0$.
This matters because if the terms are shrinking toward $0$, the partial sums tend to settle down instead of bouncing wildly. That is why the series can have a finite sum even though it never actually stops.
A partial sum is the sum of the first $N$ terms:
$$S_N = \sum_{n=1}^{N} (-1)^{n-1}a_n$$
The true infinite sum is often written as $S$. Since we can only compute a finite number of terms, we need a way to measure the error $|S-S_N|$.
The Alternating Series Error Bound
The key result is simple and very useful:
$$|S-S_N| \le a_{N+1}$$
This says that the error made by stopping after $N$ terms is at most the absolute value of the next term in the series. In other words, the first term you did not include gives a safe upper bound for how far your approximation might be from the true value.
This works because in an alternating series with decreasing terms, each partial sum overshoots or undershoots the true value by less and less. The exact sum lies between consecutive partial sums, so the distance from $S_N$ to $S$ cannot exceed the size of the next unused term.
A helpful way to remember it is:
- Last term included helps form the approximation.
- Next term omitted bounds the error.
That makes this test especially convenient on AP Calculus BC problems, where you may only need a guaranteed level of accuracy such as $0.001$.
Example 1: Estimating a Sum with Error Control
Consider
$$\sum_{n=1}^{\infty} \frac{(-1)^{n-1}}{n} = 1-\frac{1}{2}+\frac{1}{3}-\frac{1}{4}+\cdots$$
This is the alternating harmonic series. It converges, and the Alternating Series Error Bound applies because $\frac{1}{n}$ decreases and approaches $0$.
Suppose you use the first $4$ terms:
$$S_4 = 1-\frac{1}{2}+\frac{1}{3}-\frac{1}{4} = \frac{7}{12}$$
The error satisfies
$$|S-S_4| \le \frac{1}{5}$$
So the approximation is guaranteed to be within $\frac{1}{5}$ of the true sum.
If you want a better approximation, you need more terms. For example, using $10$ terms gives
$$|S-S_{10}| \le \frac{1}{11}$$
The more terms you add, the smaller the error bound becomes. This is why alternating series are often practical for approximation: the terms shrink in a predictable way.
Example 2: Meeting a Desired Accuracy
Suppose students wants to approximate a series so that the error is less than $0.01$. Consider
$$\sum_{n=1}^{\infty} (-1)^{n-1}\frac{1}{n^2}$$
Because $a_n=\frac{1}{n^2}$ decreases and approaches $0$, the Alternating Series Error Bound applies.
We want
$$a_{N+1} < 0.01$$
So solve
$$\frac{1}{(N+1)^2} < 0.01$$
This is equivalent to
$$(N+1)^2 > 100$$
so
$$N+1 > 10$$
Thus $N \ge 10$ works. Using the first $10$ terms guarantees an error less than $0.01$.
This type of question is common on AP exams: you are not just asked to approximate a sum, but also to justify how many terms are needed for a certain accuracy. The error bound gives a quick, reliable method.
Why the Next Term Gives the Error Bound
The logic behind the bound is built on the way partial sums behave. In an alternating series with decreasing term size, the partial sums move back and forth around the true sum. One partial sum is a little too high, the next is a little too low, and the gap between them is exactly the size of the next term.
For example, if the series is
$$a_1-a_2+a_3-a_4+\cdots$$
then the difference between consecutive partial sums is
$$S_{N+1}-S_N = (-1)^N a_{N+1}$$
so the size of that jump is $a_{N+1}$. Since the true sum lies between those two partial sums, the error from using $S_N$ cannot be bigger than that jump.
This explains why the error bound is not random; it comes directly from the structure of alternating signs and decreasing term sizes. It is a special convergence tool, not a general rule for all series.
Common Mistakes to Avoid
A few errors show up often in AP Calculus BC problems:
- Forgetting to check decreasing terms: The bound only works if $a_n$ decreases eventually.
- Ignoring the limit condition: You also need $\lim_{n\to\infty} a_n = 0$.
- Using the wrong term for the error: The error is bounded by the first omitted term, which is $a_{N+1}$, not $a_N$.
- Confusing the actual error with the bound: The error may be smaller than $a_{N+1}$, but the bound guarantees it will not be larger.
- Assuming the series must be positive: Alternating series can sum to positive or negative values; the alternating pattern is about signs, not the final answer.
A useful AP strategy is to write your justification in words and symbols. For example: “Since $a_n$ decreases and $\lim_{n\to\infty} a_n=0$, the Alternating Series Error Bound applies. Therefore, $|S-S_N|\le a_{N+1}$.” That short explanation is often enough for full credit when paired with correct calculations.
Connection to Taylor and Maclaurin Series
Alternating Series Error Bound also connects strongly to Taylor and Maclaurin series, which are a major AP Calculus BC topic. Many Taylor series have alternating signs, such as the Maclaurin series for $\ln(1+x)$ on certain intervals or the series for $\arctan(x)$.
When a function is approximated by a Taylor polynomial, the error is called the remainder or truncation error. For alternating Taylor series, the Alternating Series Error Bound often gives a simple estimate:
$$|R_N(x)| \le a_{N+1}$$
where $R_N(x)$ is the remainder after $N$ terms.
This is powerful because it lets you stop after a manageable number of terms and still know how accurate your approximation is. For instance, if a calculator is not allowed or exact computation is difficult, a Taylor approximation plus an error bound gives a mathematically justified estimate.
Conclusion
students, the Alternating Series Error Bound is one of the most practical ideas in infinite series 📘. It tells you that for a convergent alternating series, the error after stopping at $N$ terms is no more than the magnitude of the next term:
$$|S-S_N| \le a_{N+1}$$
This makes it possible to estimate sums, choose enough terms for a desired accuracy, and understand how alternating series behave. It also connects directly to Taylor and Maclaurin series, where approximations and error control are essential. On AP Calculus BC, this concept helps you justify answers clearly and efficiently.
Study Notes
- An alternating series has terms whose signs switch back and forth.
- A standard form is
$$\sum_{n=1}^{\infty} (-1)^{n-1}a_n$$
where $a_n>0$.
- The Alternating Series Test says the series converges if $a_n$ decreases eventually and $\lim_{n\to\infty} a_n=0$.
- The Alternating Series Error Bound is
$$|S-S_N| \le a_{N+1}$$
- The error is bounded by the first omitted term.
- To achieve a target accuracy, choose $N$ so that
$$a_{N+1}<\text{desired error}$$
- This idea is especially useful for approximating sums and Taylor series values.
- Always check both required conditions before using the bound.
- The bound gives a guaranteed maximum error, not the exact error.
- In AP Calculus BC, a clear explanation plus correct computation is often needed for full credit.