Harmonic series: \(\sum_{n=1}^{\infty} \frac{1}{n}\) (p-series with \(p = 1\))
Integral test: If \(f(x)\) is positive, continuous, and decreasing for \(x \geq N\), then \(\sum f(n)\) and \(\int f(x)dx\) both converge or both diverge
- Identify the function \(f(x) = \frac{1}{x}\)
- Verify conditions for integral test
- Evaluate the improper integral \(\int_1^{\infty} \frac{1}{x} dx\)
- Apply the integral test conclusion
Let \(f(x) = \frac{1}{x}\) for \(x \geq 1\).
We need to check that \(f(x)\) is: positive, continuous, and decreasing for \(x \geq 1\).
- Positive: \(f(x) = \frac{1}{x} > 0\) for \(x \geq 1\)
- Continuous: \(f(x) = \frac{1}{x}\) is continuous for \(x \geq 1\)
- Decreasing: \(f'(x) = -\frac{1}{x^2} < 0\) for \(x \geq 1\)
\(\int_1^{\infty} \frac{1}{x} \, dx\)
\(\int_1^{\infty} \frac{1}{x} \, dx = \lim_{t \to \infty} \int_1^{t} \frac{1}{x} \, dx\)
\(= \lim_{t \to \infty} [\ln|x|]_1^t\)
\(= \lim_{t \to \infty} (\ln|t| - \ln|1|)\)
\(= \lim_{t \to \infty} (\ln(t) - 0)\)
\(= \lim_{t \to \infty} \ln(t) = \infty\)
Since the improper integral \(\int_1^{\infty} \frac{1}{x} \, dx\) diverges to infinity,
by the integral test, the harmonic series \(\sum_{n=1}^{\infty} \frac{1}{n}\) also diverges.
Harmonic series diverges
The harmonic series \(\sum_{n=1}^{\infty} \frac{1}{n}\) diverges.
This is proven using the integral test, which shows that \(\int_1^{\infty} \frac{1}{x} dx\) diverges.
• Integral test: Series and integral both converge or both diverge under conditions
• Improper integrals: \(\int_1^{\infty} \frac{1}{x} dx = \lim_{t \to \infty} \ln(t) = \infty\)
• Logarithmic growth: Even though terms approach 0, the series diverges
p-series: \(\sum_{n=1}^{\infty} \frac{1}{n^p}\) converges if \(p > 1\), diverges if \(p \leq 1\)
\(\sum_{n=1}^{\infty} \frac{1}{n^{1.5}}\) is a p-series with \(p = 1.5\).
For a p-series \(\sum_{n=1}^{\infty} \frac{1}{n^p}\):
- Converges if \(p > 1\)
- Diverges if \(p \leq 1\)
Since \(p = 1.5 > 1\), the series converges.
Let \(f(x) = \frac{1}{x^{1.5}} = x^{-1.5}\).
\(\int_1^{\infty} x^{-1.5} \, dx = \lim_{t \to \infty} \int_1^{t} x^{-1.5} \, dx\)
\(= \lim_{t \to \infty} \left[\frac{x^{-0.5}}{-0.5}\right]_1^t = \lim_{t \to \infty} \left[-\frac{2}{\sqrt{x}}\right]_1^t\)
\(= \lim_{t \to \infty} \left(-\frac{2}{\sqrt{t}} + 2\right) = 0 + 2 = 2\)
Since the integral converges to 2, the series converges.
The series \(\sum_{n=1}^{\infty} \frac{1}{n^{1.5}}\) converges because it is a p-series with \(p = 1.5 > 1\).
Series converges
The series \(\sum_{n=1}^{\infty} \frac{1}{n^{1.5}}\) converges.
This is a p-series with \(p = 1.5 > 1\), so it converges by the p-series test.
• p-series test: \(\sum \frac{1}{n^p}\) converges iff \(p > 1\)
• Integral test: Verifies convergence for \(p > 1\)
• Power rule: \(\int x^{-p} dx = \frac{x^{1-p}}{1-p}\) for \(p \neq 1\)
Boundary behavior: The value p = 1 is the critical threshold for p-series convergence
This is a p-series with \(p = 0.9\).
Since \(0.9 < 1\), the series diverges.
This is a p-series with \(p = 1.01\).
Since \(1.01 > 1\), the series converges.
This is a p-series with \(p = 2\).
Since \(2 > 1\), the series converges.
• \(p = 0.9\): Diverges (terms decrease too slowly)
• \(p = 1.01\): Converges (terms decrease fast enough)
• \(p = 2\): Converges (terms decrease rapidly)
The value \(p = 1\) is the critical threshold:
- For \(p \leq 1\): Terms decrease too slowly, leading to divergence
- For \(p > 1\): Terms decrease fast enough for convergence
- At \(p = 1\): The harmonic series, which is the boundary case
Larger values of \(p\) lead to faster convergence:
\(\sum \frac{1}{n^2}\) converges faster than \(\sum \frac{1}{n^{1.01}}\).
\(p = 1.01\): Converges
\(p = 2\): Converges
Critical threshold: \(p = 1\)
\(\sum_{n=1}^{\infty} \frac{1}{n^{0.9}}\) diverges
\(\sum_{n=1}^{\infty} \frac{1}{n^{1.01}}\) converges
\(\sum_{n=1}^{\infty} \frac{1}{n^2}\) converges
The value \(p = 1\) is the critical threshold for convergence.
• p-series test: Convergence depends critically on \(p > 1\)
• Rate of decay: Larger p means faster decay and faster convergence
• Critical threshold: \(p = 1\) separates convergent and divergent cases
p-series: A series of the form \(\sum_{n=1}^{\infty} \frac{1}{n^p}\) where p is a positive real number
Harmonic series: The p-series with \(p = 1\): \(\sum_{n=1}^{\infty} \frac{1}{n}\)
Convergence condition: p-series converges if and only if \(p > 1\)
Critical threshold: The value \(p = 1\) separates convergent and divergent cases
Rate of convergence: Larger values of p lead to faster convergence
- Identify as p-series: Recognize the form \(\sum \frac{1}{n^p}\)
- Determine p value: Identify the exponent p
- Compare with 1: Check if \(p > 1\) or \(p \leq 1\)
- Apply the test: Use p-series convergence criterion
- Verify with integral test: If needed, confirm with integral test
• Convergence criterion: \(\sum \frac{1}{n^p}\) converges iff \(p > 1\)
• Harmonic series: \(\sum \frac{1}{n}\) diverges (p = 1)
• Basel problem: \(\sum \frac{1}{n^2} = \frac{\pi^2}{6}\)
• Rate comparison: Larger p values converge faster
• Integral connection: Related to the Riemann zeta function \(\zeta(p)\)
Comparison test: If \(0 \leq a_n \leq b_n\) for all n, then: if \(\sum b_n\) converges, then \(\sum a_n\) converges; if \(\sum a_n\) diverges, then \(\sum b_n\) diverges
Since the denominator is \(n^2 + 1\), we compare with \(\frac{1}{n^2}\) (p-series with \(p = 2 > 1\)).
For all \(n \geq 1\): \(n^2 + 1 > n^2\)
Therefore: \(\frac{1}{n^2 + 1} < \frac{1}{n^2}\)
So \(0 < \frac{1}{n^2 + 1} < \frac{1}{n^2}\) for all \(n \geq 1\).
We have \(0 \leq a_n \leq b_n\) where \(a_n = \frac{1}{n^2 + 1}\) and \(b_n = \frac{1}{n^2}\).
\(\sum_{n=1}^{\infty} \frac{1}{n^2}\) is a p-series with \(p = 2 > 1\), so it converges.
Since \(0 \leq \frac{1}{n^2 + 1} \leq \frac{1}{n^2}\) and \(\sum \frac{1}{n^2}\) converges,
then \(\sum_{n=1}^{\infty} \frac{1}{n^2 + 1}\) also converges.
The series \(\sum \frac{1}{n^2 + 1}\) behaves similarly to \(\sum \frac{1}{n^2}\) for large n.
Adding 1 to the denominator makes the terms smaller, preserving convergence.
\(\sum \frac{1}{n^2}\) converges
Therefore \(\sum \frac{1}{n^2 + 1}\) converges
The series \(\sum_{n=1}^{\infty} \frac{1}{n^2 + 1}\) converges.
This is proven using the comparison test with the convergent p-series \(\sum \frac{1}{n^2}\).
• Comparison test: Compare with a known convergent series
• p-series test: \(\sum \frac{1}{n^p}\) converges if \(p > 1\)
• Inequality manipulation: \(n^2 + 1 > n^2 \Rightarrow \frac{1}{n^2 + 1} < \frac{1}{n^2}\)
Limit comparison test: If \(\lim_{n \to \infty} \frac{a_n}{b_n} = L\) where \(0 < L < \infty\), then \(\sum a_n\) and \(\sum b_n\) both converge or both diverge
For large n: \(\sqrt{n^2 + 1} \approx \sqrt{n^2} = n\) and \(n^3 + 2 \approx n^3\)
So \(\frac{\sqrt{n^2 + 1}}{n^3 + 2} \approx \frac{n}{n^3} = \frac{1}{n^2}\)
Use \(b_n = \frac{1}{n^2}\) (convergent p-series with \(p = 2 > 1\))
\(\lim_{n \to \infty} \frac{a_n}{b_n} = \lim_{n \to \infty} \frac{\frac{\sqrt{n^2 + 1}}{n^3 + 2}}{\frac{1}{n^2}}\)
\(= \lim_{n \to \infty} \frac{\sqrt{n^2 + 1}}{n^3 + 2} \cdot n^2\)
\(= \lim_{n \to \infty} \frac{n^2\sqrt{n^2 + 1}}{n^3 + 2}\)
\(= \lim_{n \to \infty} \frac{n^2\sqrt{n^2(1 + \frac{1}{n^2})}}{n^3(1 + \frac{2}{n^3})}\)
\(= \lim_{n \to \infty} \frac{n^2 \cdot n\sqrt{1 + \frac{1}{n^2}}}{n^3(1 + \frac{2}{n^3})}\)
\(= \lim_{n \to \infty} \frac{n^3\sqrt{1 + \frac{1}{n^2}}}{n^3(1 + \frac{2}{n^3})}\)
\(= \lim_{n \to \infty} \frac{\sqrt{1 + \frac{1}{n^2}}}{1 + \frac{2}{n^3}}\)
\(= \frac{\sqrt{1 + 0}}{1 + 0} = \frac{1}{1} = 1\)
Since \(\lim_{n \to \infty} \frac{a_n}{b_n} = 1\) where \(0 < 1 < \infty\),
and \(\sum b_n = \sum \frac{1}{n^2}\) converges (p-series with \(p = 2 > 1\)),
then \(\sum_{n=1}^{\infty} \frac{\sqrt{n^2 + 1}}{n^3 + 2}\) also converges.
\(\sum \frac{1}{n^2}\) converges
Therefore original series converges
The series \(\sum_{n=1}^{\infty} \frac{\sqrt{n^2 + 1}}{n^3 + 2}\) converges.
This is proven using the limit comparison test with \(\sum \frac{1}{n^2}\).
• Limit comparison test: Compare with a known series using limits
• Asymptotic analysis: Focus on dominant terms for large n
• p-series test: \(\sum \frac{1}{n^p}\) converges if \(p > 1\)
Riemann zeta function: \(\zeta(p) = \sum_{n=1}^{\infty} \frac{1}{n^p}\) for \(p > 1\)
Basel problem: The famous result \(\zeta(2) = \frac{\pi^2}{6}\)
Convergence rate: How quickly partial sums approach the limit
Error estimation: Bounds on the difference between partial sums and the infinite sum
Generalized harmonic numbers: \(H_n^{(p)} = \sum_{k=1}^{n} \frac{1}{k^p}\)
- Integral test: For positive, decreasing functions
- Comparison test: Compare with known series
- Limit comparison test: Compare ratios of series
- Ratio test: \(\lim|\frac{a_{n+1}}{a_n}|\)
- Root test: \(\lim\sqrt[n]{|a_n|}\)
- Cauchy condensation test: For positive, decreasing series
• Integral test connection: \(\sum \frac{1}{n^p}\) and \(\int \frac{1}{x^p} dx\) both converge or diverge
• Rate of convergence: Larger p values converge faster
• Error bound: For \(\sum \frac{1}{n^p}\), the error after N terms is approximately \(\frac{1}{(p-1)N^{p-1}}\)
• Special values: \(\zeta(2n)\) has closed forms involving \(\pi\)
• Functional equation: Relates \(\zeta(s)\) and \(\zeta(1-s)\)
Questions & Answers
Question: Why does the harmonic series \(\sum \frac{1}{n}\) diverge even though the terms approach 0?
Answer: This is a common source of confusion:
- Necessary condition: For convergence, terms must approach 0 (but this isn't sufficient)
- Rate of decay: Terms must approach 0 "fast enough" for the sum to be finite
- Harmonic series: \(\frac{1}{n}\) approaches 0 too slowly to make the sum finite
Intuitively, even though each term gets smaller, they don't shrink fast enough:
- The terms \(\frac{1}{n}\) decrease logarithmically in their cumulative effect
- Groups of terms like \(\frac{1}{2} + \frac{1}{3} + \frac{1}{4}\) still contribute significantly
- Compare to \(\sum \frac{1}{n^2}\) which decreases fast enough to converge
The harmonic series is the "boundary case" where terms decrease too slowly.
Question: How do I determine if a series is a p-series when it's not in the standard form?
Answer: Look for the characteristic form by algebraic manipulation:
Standard p-series: \(\sum \frac{1}{n^p}\)
Recognizable forms:
- \(\sum \frac{1}{n^{3/2}}\) (p = 3/2)
- \(\sum \frac{1}{\sqrt{n}} = \sum \frac{1}{n^{1/2}}\) (p = 1/2)
- \(\sum \frac{1}{n^{\pi}}\) (p = π)
For more complex forms, try to rewrite:
- Factor out constants
- Use exponent rules: \(\frac{1}{\sqrt[3]{n^2}} = \frac{1}{n^{2/3}}\)
- Combine terms with common bases
- Simplify radicals and powers
If you can't express it as \(\frac{1}{n^p}\), it's not a p-series.
Question: What happens when p is between 1 and 2? How fast does it converge compared to p = 2?
Answer: When p is between 1 and 2, the series converges, but more slowly:
- Convergence: Any p > 1 converges, including values like p = 1.1, p = 1.5
- Rate of convergence: Larger p values converge faster
- Comparison: \(\sum \frac{1}{n^2}\) converges faster than \(\sum \frac{1}{n^{1.5}}\)
For example:
- \(\sum \frac{1}{n^{1.1}}\) converges, but very slowly (approaches its limit gradually)
- \(\sum \frac{1}{n^{1.5}}\) converges faster than p = 1.1 but slower than p = 2
- \(\sum \frac{1}{n^2}\) converges relatively quickly
The closer p is to 1, the slower the convergence and the more terms needed for accuracy.
Question: What's the connection between p-series and the Riemann zeta function?
Answer: The Riemann zeta function is the extension of p-series:
\(\zeta(s) = \sum_{n=1}^{\infty} \frac{1}{n^s}\)
When s is a real number greater than 1, this is exactly the p-series with p = s.
Key connections:
- Domain: The series definition only works for \(\text{Re}(s) > 1\)
- Extension: The zeta function extends analytically to the entire complex plane (except s = 1)
- Special values: \(\zeta(2) = \frac{\pi^2}{6}\), \(\zeta(4) = \frac{\pi^4}{90}\), etc.
- Riemann Hypothesis: One of the most famous unsolved problems in mathematics
For real p > 1, \(\zeta(p)\) gives the sum of the corresponding p-series.
Question: How do p-series apply to practical computing and engineering?
Answer: p-series have numerous practical applications:
- Numerical methods: Understanding convergence rates of algorithms
- Probability: Modeling rare events and heavy-tailed distributions
- Physics: Power-law relationships in natural phenomena
- Computer science: Analyzing algorithms with power-law behavior
Examples:
- Error analysis: Knowing convergence rates helps estimate computational accuracy
- Network analysis: Power-law distributions model many real-world networks
- Signal processing: Understanding decay rates of frequency components
Understanding p-series helps engineers predict system behavior and computational requirements.