Alternating series test (Leibniz test): If \(\sum (-1)^{n+1} b_n\) where \(b_n > 0\) satisfies: 1) \(b_n\) decreases monotonically, 2) \(\lim_{n \to \infty} b_n = 0\), then the series converges
Alternating harmonic series: The classic example \(\sum \frac{(-1)^{n+1}}{n}\)
- Identify the form: \(\sum (-1)^{n+1} b_n\) where \(b_n > 0\)
- Verify: \(b_n > 0\) for all \(n \geq N\)
- Verify: \(b_n\) is decreasing (show \(b_{n+1} \leq b_n\))
- Verify: \(\lim_{n \to \infty} b_n = 0\)
- Apply the alternating series test conclusion
We have \(\sum_{n=1}^{\infty} \frac{(-1)^{n+1}}{n} = \sum_{n=1}^{\infty} (-1)^{n+1} \cdot \frac{1}{n}\)
So \(b_n = \frac{1}{n}\) and we need to verify the conditions for the alternating series test.
For all \(n \geq 1\): \(b_n = \frac{1}{n} > 0\)
✓ Condition satisfied: all terms are positive.
We need to show that \(b_n = \frac{1}{n}\) is decreasing.
Check: \(b_{n+1} \leq b_n\)
\(\frac{1}{n+1} \leq \frac{1}{n}\)
This is true because \(n+1 > n\) implies \(\frac{1}{n+1} < \frac{1}{n}\).
✓ Condition satisfied: sequence is decreasing.
We need: \(\lim_{n \to \infty} b_n = \lim_{n \to \infty} \frac{1}{n} = 0\)
This is a basic limit: as \(n \to \infty\), \(\frac{1}{n} \to 0\).
✓ Condition satisfied: limit is zero.
Since all conditions are satisfied:
- \(b_n = \frac{1}{n} > 0\) for all \(n \geq 1\)
- \(b_n\) is decreasing: \(\frac{1}{n+1} < \frac{1}{n}\)
- \(\lim_{n \to \infty} b_n = 0\)
By the alternating series test, \(\sum_{n=1}^{\infty} \frac{(-1)^{n+1}}{n}\) converges.
Decreasing: ✓
Limit: 0 ✓
Converges
The series \(\sum_{n=1}^{\infty} \frac{(-1)^{n+1}}{n}\) converges.
This is proven using the alternating series test with \(b_n = \frac{1}{n}\).
• Alternating series test: Three conditions must be verified
• Monotonicity verification: Show \(b_{n+1} \leq b_n\) directly
• Basic limit: \(\lim_{n \to \infty} \frac{1}{n} = 0\)
Rational alternating series: Alternating series with rational function terms
We have \(\sum_{n=1}^{\infty} \frac{(-1)^{n+1}}{n^2 + 1} = \sum_{n=1}^{\infty} (-1)^{n+1} \cdot \frac{1}{n^2 + 1}\)
So \(b_n = \frac{1}{n^2 + 1}\).
For all \(n \geq 1\): \(b_n = \frac{1}{n^2 + 1} > 0\)
Since \(n^2 + 1 > 0\) for all real n, we have \(b_n > 0\).
✓ Condition satisfied.
We need to show that \(b_n = \frac{1}{n^2 + 1}\) is decreasing.
Check: \(b_{n+1} \leq b_n\)
\(\frac{1}{(n+1)^2 + 1} \leq \frac{1}{n^2 + 1}\)
This is equivalent to: \((n+1)^2 + 1 \geq n^2 + 1\)
\(n^2 + 2n + 1 + 1 \geq n^2 + 1\)
\(n^2 + 2n + 2 \geq n^2 + 1\)
\(2n + 2 \geq 1\)
\(2n \geq -1\)
Since \(n \geq 1\), we have \(2n \geq 2 > -1\), so the inequality holds.
✓ Condition satisfied: sequence is decreasing.
We need: \(\lim_{n \to \infty} \frac{1}{n^2 + 1}\)
As \(n \to \infty\), \(n^2 + 1 \to \infty\), so \(\frac{1}{n^2 + 1} \to 0\).
✓ Condition satisfied: limit is zero.
Since all conditions are satisfied, the series converges.
All conditions met
Series converges
The series \(\sum_{n=1}^{\infty} \frac{(-1)^{n+1}}{n^2 + 1}\) converges.
This is proven using the alternating series test.
• Monotonicity verification: Show inequality by cross-multiplication
• Rational function limits: \(\lim \frac{1}{p(n)} = 0\) for polynomial \(p(n)\) with positive leading coefficient
• Alternating series test: Three conditions must hold
Divergent alternating series: Series that fails one or more conditions of the alternating series test
We have \(\sum_{n=1}^{\infty} \frac{(-1)^n \cdot n}{n + 1} = \sum_{n=1}^{\infty} (-1)^n \cdot \frac{n}{n + 1}\)
So \(b_n = \frac{n}{n + 1}\).
First, let's find \(\lim_{n \to \infty} b_n = \lim_{n \to \infty} \frac{n}{n + 1}\)
Divide numerator and denominator by n:
\(\lim_{n \to \infty} \frac{1}{1 + \frac{1}{n}} = \frac{1}{1 + 0} = 1\)
Since \(\lim_{n \to \infty} b_n = 1 \neq 0\), the limit condition fails.
Since \(\lim_{n \to \infty} \frac{(-1)^n \cdot n}{n + 1}\) does not exist (oscillates between ±1),
the series diverges by the nth term test.
For completeness, let's check if \(b_n = \frac{n}{n + 1}\) is decreasing:
\(b_{n+1} = \frac{n+1}{n+2}\) and \(b_n = \frac{n}{n+1}\)
\(\frac{n+1}{n+2} \geq \frac{n}{n+1}\) (since \((n+1)^2 \geq n(n+2)\), i.e., \(n^2 + 2n + 1 \geq n^2 + 2n\), i.e., \(1 \geq 0\))
So actually \(b_{n+1} \geq b_n\), meaning the sequence is increasing, not decreasing.
Both the limit condition and the decreasing condition fail, so the series diverges.
Limit condition fails
Series diverges
The series \(\sum_{n=1}^{\infty} \frac{(-1)^n \cdot n}{n + 1}\) diverges.
The primary reason is that \(\lim_{n \to \infty} \frac{n}{n + 1} = 1 \neq 0\), violating the limit condition of the alternating series test.
• nth term test: If \(\lim a_n \neq 0\), then \(\sum a_n\) diverges
• Alternating series test: All three conditions must be satisfied
• Limit evaluation: Divide by highest power to evaluate rational limits
Alternating series: A series of the form \(\sum (-1)^n b_n\) or \(\sum (-1)^{n+1} b_n\) where \(b_n > 0\)
Leibniz test: Another name for the alternating series test
Conditional convergence: Series converges but \(\sum |a_n|\) diverges
Alternating harmonic series: \(\sum \frac{(-1)^{n+1}}{n}\) converges conditionally
Error bound: For convergent alternating series, \(|R_N| \leq b_{N+1}\)
- Identify alternating form: Ensure series has form \(\sum (-1)^n b_n\) with \(b_n > 0\)
- Check positivity: Verify \(b_n > 0\) for all \(n \geq N\)
- Check monotonicity: Verify \(b_n\) is decreasing
- Check limit: Verify \(\lim_{n \to \infty} b_n = 0\)
- Apply conclusion: If all conditions hold, series converges
• Alternating Series Test: If \(b_n > 0\), \(b_n \searrow\), and \(\lim b_n = 0\), then \(\sum (-1)^{n+1} b_n\) converges
• Error Bound: \(|R_N| \leq b_{N+1}\) for alternating series
• Conditional vs Absolute: Alternating harmonic series converges conditionally
• Dirichlet's Test: Generalization of alternating series test
• Abel's Test: For series of the form \(\sum a_n b_n\)
Error bound: For a convergent alternating series \(\sum (-1)^{n+1} b_n\), the error after N terms is bounded by \(|R_N| \leq b_{N+1}\)
For \(b_n = \frac{1}{n^3}\):
- Positive: \(b_n = \frac{1}{n^3} > 0\) for \(n \geq 1\)
- Decreasing: \(b_{n+1} = \frac{1}{(n+1)^3} < \frac{1}{n^3} = b_n\)
- Limit: \(\lim_{n \to \infty} \frac{1}{n^3} = 0\)
All conditions are satisfied, so the series converges by AST.
For the alternating series \(\sum_{n=1}^{\infty} \frac{(-1)^{n+1}}{n^3}\), the error after N terms is bounded by:
\(|R_N| \leq b_{N+1} = \frac{1}{(N+1)^3}\)
We want: \(\frac{1}{(N+1)^3} < 0.001\)
Or: \((N+1)^3 > \frac{1}{0.001} = 1000\)
\(N+1 > \sqrt[3]{1000} = 10\)
So \(N > 9\)
Therefore, \(N \geq 10\).
With \(N = 10\): \(b_{11} = \frac{1}{11^3} = \frac{1}{1331} \approx 0.000751 < 0.001\) ✓
With \(N = 9\): \(b_{10} = \frac{1}{10^3} = 0.001\) (not strictly less than 0.001)
So we need at least 10 terms.
\(N \geq 10\)
Use first 10 terms
We need at least 10 terms to approximate the sum with an error less than 0.001.
The error bound ensures that using the first 10 terms gives an approximation accurate to within 0.001.
• Error bound theorem: \(|R_N| \leq b_{N+1}\) for alternating series
• Monotonicity verification: Show terms are decreasing
• Cube root calculation: \(\sqrt[3]{1000} = 10\)
Trigonometric alternating series: Series with both alternating signs and trigonometric functions
First, let's examine \(\cos(n\pi/2)\) for different values of n:
- \(n = 1\): \(\cos(\pi/2) = 0\)
- \(n = 2\): \(\cos(\pi) = -1\)
- \(n = 3\): \(\cos(3\pi/2) = 0\)
- \(n = 4\): \(\cos(2\pi) = 1\)
- \(n = 5\): \(\cos(5\pi/2) = 0\)
- \(n = 6\): \(\cos(3\pi) = -1\)
Pattern: \(\cos(n\pi/2) = 0\) when n is odd, and \(\cos(n\pi/2) = (-1)^{n/2}\) when n is even.
Since \(\cos(n\pi/2) = 0\) for odd n, only even terms contribute:
\(\sum_{n=1}^{\infty} \frac{(-1)^n \cos(n\pi/2)}{n^2 + 1} = \sum_{k=1}^{\infty} \frac{(-1)^{2k} \cos(k\pi)}{(2k)^2 + 1}\)
\(= \sum_{k=1}^{\infty} \frac{1 \cdot (-1)^k}{(2k)^2 + 1}\)
\(= \sum_{k=1}^{\infty} \frac{(-1)^k}{4k^2 + 1}\)
Now we have \(\sum_{k=1}^{\infty} \frac{(-1)^k}{4k^2 + 1}\) with \(b_k = \frac{1}{4k^2 + 1}\).
Check conditions:
- Positive: \(b_k = \frac{1}{4k^2 + 1} > 0\) for all \(k \geq 1\)
- Decreasing: \(b_{k+1} = \frac{1}{4(k+1)^2 + 1} < \frac{1}{4k^2 + 1} = b_k\)
- Limit: \(\lim_{k \to \infty} \frac{1}{4k^2 + 1} = 0\)
To show \(b_{k+1} < b_k\):
\(4(k+1)^2 + 1 > 4k^2 + 1\)
\(4(k^2 + 2k + 1) + 1 > 4k^2 + 1\)
\(4k^2 + 8k + 4 + 1 > 4k^2 + 1\)
\(8k + 4 > 0\)
This is true for all \(k \geq 1\).
Since all conditions of the alternating series test are satisfied,
the series \(\sum_{k=1}^{\infty} \frac{(-1)^k}{4k^2 + 1}\) converges.
All AST conditions met
Converges
The series \(\sum_{n=1}^{\infty} \frac{(-1)^n \cos(n\pi/2)}{n^2 + 1}\) converges.
This is proven by recognizing that only even-indexed terms contribute, reducing it to an alternating series that satisfies the AST.
• Trigonometric pattern recognition: \(\cos(n\pi/2)\) has zeros at odd indices
• Series reduction: Eliminate zero terms to simplify
• Alternating series test: Verify all three conditions
Conditional convergence: A series converges but the series of absolute values diverges
Abel's test: For series of the form \(\sum a_n b_n\) where \(\sum a_n\) converges and \(\{b_n\}\) is monotonic and bounded
Dirichlet's test: Generalization of alternating series test for more general oscillating factors
Riemann series theorem: Conditionally convergent series can be rearranged to converge to any value
Uniform convergence: For sequences of functions, important for power series
- Asymptotic analysis: Focus on dominant terms for large n
- Comparison with known series: Use established convergent alternating series
- Integral representation: Express terms using integrals for analysis
- Generating functions: Transform series into functions
- Fourier analysis: Use harmonic analysis techniques
- Complex analysis: Extend to complex plane for advanced results
• Riemann Series Theorem: Any conditionally convergent series can be rearranged to converge to any desired value
• Abel's Test: If \(\sum a_n\) converges and \(\{b_n\}\) is monotonic and bounded, then \(\sum a_n b_n\) converges
• Dirichlet's Test: If partial sums of \(\{a_n\}\) are bounded and \(\{b_n\} \searrow 0\), then \(\sum a_n b_n\) converges
• Uniform Convergence: For series of functions, important for preserving continuity and differentiability
• Abel Summation: Technique for proving convergence of certain series
Questions & Answers
Question: What's the difference between absolute and conditional convergence? How does this relate to the alternating series test?
Answer: These are fundamental concepts in series theory:
- Absolute convergence: \(\sum |a_n|\) converges (implies \(\sum a_n\) converges)
- Conditional convergence: \(\sum a_n\) converges but \(\sum |a_n|\) diverges
The alternating series test can only prove conditional convergence:
- AST proves \(\sum (-1)^n b_n\) converges when conditions are met
- But \(\sum |(-1)^n b_n| = \sum b_n\) might diverge
- Example: \(\sum \frac{(-1)^n}{n}\) converges (by AST) but \(\sum \frac{1}{n}\) diverges
Absolutely convergent series are "more robust" - they converge regardless of term arrangement, while conditionally convergent series can be rearranged to converge to any value.
Question: How do I verify that a sequence is decreasing? Is there a general method?
Answer: There are several methods to verify that \(b_n\) is decreasing:
- Direct comparison: Show \(b_{n+1} \leq b_n\) by algebraic manipulation
- Derivative test: If \(f(x)\) is continuous and \(f'(x) < 0\) for \(x \geq N\), then \(f(n)\) is decreasing
- Ratio test: Show \(\frac{b_{n+1}}{b_n} < 1\) for all \(n \geq N\)
- Difference test: Show \(b_n - b_{n+1} > 0\)
Example: For \(b_n = \frac{1}{n^2}\), we can show:
- Direct: \(\frac{1}{(n+1)^2} < \frac{1}{n^2}\) since \((n+1)^2 > n^2\)
- Derivative: \(f(x) = \frac{1}{x^2}\), \(f'(x) = -\frac{2}{x^3} < 0\) for \(x > 0\)
Choose the method that's most convenient for the given sequence.
Question: What happens if only two of the three AST conditions are satisfied?
Answer: The alternating series test requires ALL THREE conditions:
- \(b_n > 0\) for all \(n \geq N\)
- \(b_n\) is decreasing
- \(\lim_{n \to \infty} b_n = 0\)
If any condition fails:
- Positive condition fails: Not an alternating series of the required form
- Decreasing condition fails: AST doesn't apply, need other tests
- Limit condition fails: Series diverges by nth term test
Example: \(\sum \frac{(-1)^n n}{n+1}\) fails the limit condition since \(\lim \frac{n}{n+1} = 1 \neq 0\), so it diverges.
Question: Can the alternating series test prove divergence?
Answer: The alternating series test is ONLY a convergence test:
- If all conditions are satisfied → series converges
- If conditions are not satisfied → test is inconclusive (series may converge or diverge)
To prove divergence, use:
- Nth term test: If \(\lim a_n \neq 0\) or doesn't exist, then \(\sum a_n\) diverges
- Comparison tests: Compare with known divergent series
- Other tests: Ratio test, root test, etc.
Example: \(\sum \frac{(-1)^n}{\sqrt{n}}\) satisfies AST conditions and converges, but \(\sum \frac{(-1)^n n}{n+1}\) fails the limit condition and diverges by the nth term test.
Question: How are alternating series used in practical applications?
Answer: Alternating series have numerous practical applications:
- Numerical analysis: Taylor series often have alternating signs, allowing precise error bounds
- Signal processing: Fourier series use alternating patterns for signal representation
- Physics: Perturbation methods in quantum mechanics use alternating series
- Computer science: Algorithms for numerical computation rely on alternating series
Example: The series \(\ln(2) = \sum \frac{(-1)^{n+1}}{n}\) is used in computational algorithms for logarithm calculation, with the error bound helping determine how many terms are needed for desired accuracy.