📝 Convergence test for series (36 MCQs)

📖 Explanation: This question targets a common conceptual misconception regarding hierarchy of growth. While it is true that exponentials dominate polynomials, factorials ($k!$) grow significantly faster than geometric terms ($b^k$). Applying the Ratio Test confirms this: the limit of successive terms is $\lim_{k \to \infty} \frac{(k+1)!}{3^{k+1}} \cdot \frac{3^k}{k!} = \lim_{k \to \infty} \frac{k+1}{3} = \infty$. Since the limit exceeds 1, the series diverges. Understanding this hierarchy is crucial for selecting appropriate tests without unnecessary computation.

📖 Explanation: The Direct Comparison Test requires the inequality to hold for all terms (or eventually), but finding the specific crossover point can be algebraically tedious. The Limit Comparison Test bypasses this by focusing solely on asymptotic behavior. Computing $\rho = \lim_{k \to \infty} \frac{a_k}{b_k} = \lim_{k \to \infty} \frac{2k^2+5}{k^4-k+1} \cdot k^2 = 2$. Since $0 < \rho < \infty$ and $\sum \frac{1}{k^2}$ converges, the original series converges. This highlights the strategic advantage of limit-based methods over direct inequalities for rational functions.

📖 Explanation: In physical modeling, quantities like energy are scalar and additive; their total value should not depend on the arbitrary order of summation. Riemann's Rearrangement Theorem states that conditionally convergent series can be rearranged to sum to any value, which violates physical conservation laws. By establishing absolute convergence via $\sum |\frac{\sin k}{k^2}| \leq \sum \frac{1}{k^2}$, we ensure the model is robust and well-defined. This connects mathematical rigor to physical validity, emphasizing why absolute convergence is often a stricter requirement in applied sciences than mere convergence.

📖 Explanation: This represents a fundamental logical error known as affirming the consequent. The Divergence Test states: if $\sum a_k$ converges, then $\lim a_k = 0$. The converse is false. Many divergent series, including p-series with $p \leq 1$, have terms approaching zero. The harmonic series and $\sum \frac{1}{\sqrt{k}}$ are classic counterexamples. Students must understand that $\lim a_k = 0$ is a necessary but not sufficient condition for convergence. Recognizing this distinction prevents misapplication of the most basic convergence tool.

📖 Explanation: Graphical evidence of partial sums is suggestive but never proof. A series like $\sum \frac{1}{k \ln k}$ diverges incredibly slowly, appearing flat on standard scales even after millions of terms. Conversely, a convergent series might appear to grow linearly before suddenly leveling off. Without analytical bounds or a closed form, visual flattening is insufficient to distinguish between very slow divergence and convergence. This question reinforces the necessity of analytical tests over numerical or graphical intuition, especially for borderline cases where computational resources cannot reveal asymptotic behavior.

📖 Explanation: When terms involve expressions raised to the $k$-th power, the Root Test is often more effective than the Ratio Test. Here, $\rho = \lim_{k \to \infty} \sqrt[k]{|a_k|} = \lim_{k \to \infty} \frac{2k+1}{3k-5} = \frac{2}{3}$. Since $\rho < 1$, the series converges absolutely. The Ratio Test would require evaluating $\lim \frac{a_{k+1}}{a_k}$, leading to complex indeterminate forms involving powers. This problem tests strategic selection: recognizing structural cues (powers vs. factorials) that dictate optimal test choice, saving significant computational effort and avoiding dead ends.

📖 Explanation: Numerical approximation cannot distinguish between absolute and conditional convergence. The given series converges by the Alternating Series Test since $\frac{k}{k^2+1}$ decreases to 0. However, $\sum |a_k| = \sum \frac{k}{k^2+1}$ behaves like $\sum \frac{1}{k}$ and diverges by Limit Comparison. Thus, the series is conditionally convergent. Computational tools display partial sums approaching a limit, masking the divergence of absolute values. This highlights a critical limitation of numerical methods: they verify existence of a sum but not its robustness under rearrangement or absolute summability.

📖 Explanation: Only option B produces absolute convergence. Option C converges conditionally, not absolutely. Options A and D still diverge (A by integral test, D by limit comparison with harmonic). The key insight is that convergence of positive-term series depends entirely on decay rate. Adding constants or logarithms doesn't change the fundamental $1/k$ decay class. Only increasing the power beyond 1 changes the integral $\int x^{-p} dx$ from logarithmic divergence to finite convergence. This tests deep understanding of threshold behavior in p-series versus superficial modifications.

📖 Explanation: This is a pervasive misconception: equating the value of the improper integral with the series sum. The Integral Test provides a binary outcome (converge/diverge) based on whether the area under the curve is finite. The series sum and integral value are generally unequal; for this series, the sum is approximately 2.1 while the integral is ~1.44. The integral serves as a bound or comparison tool, not an evaluation method. Understanding this distinction prevents erroneous quantitative claims from qualitative tests.

📖 Explanation: This tests understanding of stability under perturbation. Since $|\sin k| \leq 1$, we have $\frac{1}{k^2+1} \leq b_k \leq \frac{1}{k^2-1}$ for large k. Both bounding series converge (by limit comparison with $\sum 1/k^2$), so B converges by Squeeze/Comparison. More conceptually, absolutely convergent series are stable under bounded additive perturbations in the denominator. The oscillation doesn't affect the dominant $k^2$ decay rate. This avoids mechanical testing and relies on asymptotic dominance principles, demonstrating higher-order synthesis of comparison concepts.

📖 Explanation: The student's heuristic fails because $k^k$ is super-exponential, growing much faster than $k!$. Using Stirling's formula $k! \sim \sqrt{2\pi k}(k/e)^k$, the term becomes $\sim \sqrt{2\pi k}/e^k$, which decays exponentially. Ratio Test confirms: $\lim \frac{(k+1)!}{(k+1)^{k+1}} \cdot \frac{k^k}{k!} = \lim (1+\frac{1}{k})^{-k} = 1/e < 1$. This Olympiad-style problem challenges oversimplified growth hierarchies and rewards precise asymptotic reasoning over memorized rules. It demonstrates that context (what the factorial is compared to) determines convergence.

📖 Explanation: Identifying dominant terms is key. Denominator has $k!$ which dominates $3^k$; numerator has $2^k$ dominating $k^3$. The series behaves like $\sum \frac{2^k}{k!}$. Ratio Test on this simplified form gives $\lim \frac{2^{k+1}}{(k+1)!} \cdot \frac{k!}{2^k} = \lim \frac{2}{k+1} = 0 < 1$. Formal application to original series yields same result. This tests ability to mentally simplify complex expressions before selecting tools, avoiding unnecessary algebra. Recognizing factorial dominance immediately signals Ratio Test applicability and predicts convergence.

📖 Explanation: By definition, conditional convergence means $\sum a_k$ converges but $\sum |a_k|$ diverges. Riemann's Rearrangement Theorem then guarantees terms can be reordered to sum to any prescribed value or diverge. This profound result distinguishes conditional from absolute convergence. Options B and C describe absolute convergence properties. Option D contradicts the definition. This question assesses theoretical understanding of series pathology rather than computational skill, emphasizing why absolute convergence is preferred in analysis and applications where order-independence is required.

📖 Explanation: CAS may fail with oscillatory non-alternating terms. Analytical verification uses comparison: $|\cos k| \leq 1 \implies |a_k| \leq 1/k^2$. Since $\sum 1/k^2$ converges (p=2), the original series converges absolutely, hence converges. The CAS likely mishandled the irregular sign pattern or used an inappropriate test. This scenario emphasizes human oversight of computational tools and reinforces that bounded oscillation combined with sufficient decay guarantees convergence. It also illustrates why absolute convergence tests are more robust for irregularly signed series than alternating-specific methods.

📖 Explanation: When Ratio Test gives 1, terms decay too slowly for geometric comparison. This series involves central binomial coefficients $\binom{2k}{k} \sim \frac{4^k}{\sqrt{\pi k}}$. Substituting yields $a_k \sim \frac{1}{\sqrt{\pi k}}$, which diverges like harmonic p=1/2 series. Stirling's formula resolves the indeterminate ratio by revealing hidden algebraic decay. Root Test also gives 1. Integral Test is impractical. This Olympiad-level problem requires specialized asymptotic knowledge beyond standard tests, testing depth of understanding of combinatorial series behavior at convergence boundaries.

📖 Explanation: Taking logs transforms multiplicative dynamics to additive: $\ln P_n = \ln P_0 + \sum \ln(1+1/k)$. Since $\ln(1+x) \sim x$ as $x \to 0$, and $\sum 1/k$ diverges, the log-sum diverges to infinity, implying $P_n \to \infty$. This connects series convergence to dynamical systems stability. The equivalence relies on continuity of ln and asymptotic equivalence. This applied scenario demonstrates how series tests inform real-world model predictions, bridging abstract math and scientific modeling through functional transformations.

📖 Explanation: Option A: both harmonic series diverge, but their sum is identically zero, which trivially converges. This illustrates that divergence is not preserved under addition when cancellation occurs. Option B sums to divergent series. Option C terms don't approach zero. Option D diverges to negative infinity. This counterintuitive result warns against assuming algebraic closure properties for divergent series. It underscores why convergence tests must be applied to combined expressions directly rather than inferring from components, especially in linear combinations arising in differential equations or signal processing.

📖 Explanation: While $(k+1)^2/k^2 \to 1$ is correct, the student missed the factorial contribution. Full ratio: $\frac{(k+1)^2 2^{k+1}}{(k+1)!} \cdot \frac{k!}{k^2 2^k} = 2 \cdot \frac{(k+1)^2}{k^2} \cdot \frac{1}{k+1} = 2 \cdot \frac{k+1}{k^2} \to 0$. The critical error was omitting $\frac{k!}{(k+1)!} = \frac{1}{k+1}$. Factorial decay dominates polynomial-exponential growth. This error analysis question teaches careful bookkeeping in limit evaluations and reinforces that factorials introduce essential $1/k$ factors that drastically alter convergence outcomes.

📖 Explanation: Decreasing oscillation amplitude suggests convergence (possibly alternating-type). But if envelope decays as $n^{-1/2}$, the absolute series behaves like $\sum n^{-1/2}$, which diverges (p=1/2 < 1). Thus, convergence is likely conditional. Graphs of partial sums reveal convergence rate visually; power-law envelopes indicate p-series behavior. Absolute convergence requires envelope decay faster than $1/n$. This visual-analytical bridge helps diagnose convergence type from numerical output, useful when closed forms are unavailable.

📖 Explanation: This is cosh(x) series. For x=10, terms grow then shrink. Remainder Estimation Theorem gives $|R_n| \leq M \frac{|x|^{n+1}}{(n+1)!}$ where M bounds next derivative. Since derivatives are sinh/cosh, M=cosh(10)≈11013. Solve $11013 \cdot 10^{n+1}/(n+1)! < 10^{-6}$. Trial shows n≈15 suffices. Alternating test doesn't apply (all positive). This applied problem combines series identification, error estimation, and practical computation planning, reflecting real scientific workflow where theoretical bounds guide numerical implementation.

📖 Explanation: As $k \to \infty$, $\arctan(k) \to \pi/2$, a positive constant. Thus $a_k \sim \frac{\pi/2}{k^3}$. By LCT with $b_k = 1/k^3$, $\rho = \pi/2$, so both converge. Choosing $1/k^2$ would work but isn't tightest; $1/k^4$ is wrong since arctan doesn't decay. Preserving arctan complicates unnecessarily. This tests recognition of bounded function asymptotics: when a factor approaches a nonzero constant, it doesn't affect convergence class. Essential skill for handling transcendental functions in series.

📖 Explanation: The Ratio Test limit $\rho$ indicates asymptotic geometric ratio. $\rho=0.999$ means terms eventually decay like $0.999^k$, which is slow but still exponential. However, 'proximity to boundary' is misleading; convergence speed also depends on pre-asymptotic behavior and constant factors. A series with $\rho=0.5$ could have huge initial terms making early partial sums worse. The test gives asymptotic guarantee, not uniform speed metric. This clarifies that $\rho$ is a limit property, not a global performance indicator, preventing misinterpretation of numerical test outputs.

📖 Explanation: Uniform convergence requires bounding |terms| by convergent numerical series independent of x. Since $|e^{-k^2 t} \sin(kx)| \leq e^{-k^2 t}$ and $\sum e^{-k^2 t}$ converges (by Ratio Test or comparison to geometric), Weierstrass M-test applies. This ensures continuity/differentiability of solution. Other tests establish pointwise convergence only. This bridges series tests and PDE theory, showing how convergence concepts extend to function spaces. Essential for validating series solutions in physics where term-by-term operations are needed.

📖 Explanation: Telescoping relies on exact algebraic cancellation $\frac{1}{k(k+1)} = \frac{1}{k} - \frac{1}{k+1}$. Adding $\sin(k)$ destroys this identity. While the perturbed series still converges (by LCT with $1/k^2$), its sum differs from 1. Numerical similarity doesn't imply identical sums. This highlights that convergence and sum value are distinct properties; structural methods like telescoping are fragile under perturbation, unlike robust comparison tests. Teaches caution in extrapolating exact results to approximate models.

📖 Explanation: Denominator is double factorial: $(2k+1)!! = \frac{(2k+1)!}{2^k k!}$. Thus $a_k = \frac{k! 2^k k!}{(2k+1)!} = \frac{2^k (k!)^2}{(2k+1)!}$. Now Ratio Test: $\frac{a_{k+1}}{a_k} = \frac{2^{k+1}((k+1)!)^2}{(2k+3)!} \cdot \frac{(2k+1)!}{2^k (k!)^2} = 2 \cdot \frac{(k+1)^2}{(2k+3)(2k+2)} \to \frac{2}{4} = 0.5 < 1$. Converges. This tests algebraic manipulation skills to convert obscure products to standard factorials, enabling routine tests. Essential for advanced series encountered in special functions.

📖 Explanation: If remainder $R_n \sim C/n^2$, then $a_{n+1} = S_{n+1} - S_n \approx R_n - R_{n+1} \sim C(1/n^2 - 1/(n+1)^2) \sim 2C/n^3$. So terms decay cubically, implying absolute convergence (p=3>1). Quadratic remainder decay indicates faster-than-harmonic term decay. This inverse relationship between remainder rate and term decay links numerical observation to analytical properties. Useful for diagnosing series behavior from computational output when closed form unknown.

📖 Explanation: Term has form $f(k)^{g(k)}$. Root Test: $(a_k)^{1/k} = \left( \frac{k+1}{2k} \right)^k \to (1/2)^k \to 0 < 1$. Converges. Ratio Test requires $\frac{a_{k+1}}{a_k} = \left( \frac{k+2}{2k+2} \right)^{(k+1)^2} / \left( \frac{k+1}{2k} \right)^{k^2}$, extremely complex. Root Test extracts exponent cleanly. This exemplifies when Root Test shines: nested exponents. Strategic test selection based on expression structure saves immense effort. Reinforces that Root Test isn't just backup—it's primary for certain forms.

📖 Explanation: Terms behave as $k e^{-c k^2}$. Exponential quadratic decay overwhelms any polynomial growth. Ratio Test: $\frac{(k+1)e^{-c(k+1)^2}}{k e^{-c k^2}} \sim e^{-2ck} \to 0$. Converges absolutely for any c>0 (i.e., T>0). Physical significance: partition function always well-defined for systems with unbounded energy spectrum growing faster than log. Connects series convergence to thermodynamic stability. Demonstrates how physical constraints (energy growth) map to mathematical convergence criteria.

📖 Explanation: Integral Test demands f(x) be positive, continuous, and DECREASING. $\sin^2 x / x^2$ oscillates, so test inapplicable. Correct approach: comparison $\sin^2 k / k^2 \leq 1/k^2$, converges. Non-elementary antiderivative doesn't imply divergence (e.g., $e^{-x^2}$). This tests precise recall of test conditions, not just mechanical application. Many forget monotonicity requirement. Reinforces that test hypotheses are gatekeepers; violating them invalidates conclusions regardless of integral computability.

📖 Explanation: Comparison Test for divergence requires LOWER bound by divergent series. Being upper-bounded by divergent series is inconclusive (could converge or diverge). Here, $\ln(k+1) \sim \ln k$, so series behaves like $\sum 1/(k \ln k)$, which diverges (integral $\int dx/(x \ln x) = \ln \ln x \to \infty$). Student's direction of inequality is backwards for proving divergence. This reinforces proper logical structure of comparison tests and asymptotic equivalence over naive inequality.

📖 Explanation: All converge, but C decays slowest. Logarithmic iterates create hierarchy: $\ln \ln k$ grows slower than $\ln k$, so denominator grows slowest, terms largest. Integral test confirms: $\int \frac{dx}{x \ln x (\ln \ln x)^2} = \int \frac{du}{u^2}$ converges. This explores Bertrand series and convergence thresholds. Understanding this hierarchy reveals that convergence isn't binary but has degrees of slowness. Important for algorithm analysis and complexity theory where borderline convergence affects runtime estimates.

📖 Explanation: Gauss/Raab's test extension: if $a_{k+1}/a_k = 1 - p/k + o(1/k)$, behaves like $k^{-p}$. Here p=2>1, so converges. Derivation: $\ln a_k \sim \sum \ln(1-2/k) \sim -2 \sum 1/k \sim -2 \ln k$, so $a_k \sim k^{-2}$. This synthesizes ratio asymptotics with p-series without invoking specialized tests. Tests deep understanding of how local ratio behavior dictates global decay. Essential for research-level analysis where standard tests fail.

📖 Explanation: Absolute convergence of Fourier coefficients implies uniform convergence of series, yielding continuous bounded H(ω). In systems theory, $\sum |c_k| < \infty$ is sufficient condition for BIBO (bounded-input bounded-output) stability. Links mathematical absolute convergence to engineering stability criteria. Shows why absolute (not conditional) convergence matters in applications: ensures robustness and well-behaved frequency domain representation. Bridges pure analysis and applied signal processing.

📖 Explanation: Let $b_k = \frac{1}{\sqrt{k} + (-1)^k}$. Check monotonicity: $b_{2m} = \frac{1}{\sqrt{2m}+1}$, $b_{2m+1} = \frac{1}{\sqrt{2m+1}-1}$. For large m, $b_{2m+1} > b_{2m}$, violating decrease. AST fails. In fact, expanding $b_k = \frac{1}{\sqrt{k}} \mp \frac{1}{k} + ...$ reveals hidden harmonic component causing divergence. This notorious counterexample shows AST hypotheses are essential, not technicalities. Tests deep understanding beyond surface pattern matching.

📖 Explanation: Algebraic simplification: $\frac{k^k}{(k+1)^{k+1}} = \frac{k^k}{(k+1)(k+1)^k} = \frac{1}{k+1} \left( \frac{k}{k+1} \right)^k = \frac{1}{k+1} \left(1+\frac{1}{k}\right)^{-k} \to \frac{1}{e(k+1)}$. Since $\sum \frac{1}{k+1}$ diverges, original diverges. Ratio/Root Tests give 1/e < 1? Wait: recalculate ratio: $\frac{a_{k+1}}{a_k} = \frac{(k+1)^{k+1}}{(k+2)^{k+2}} \cdot \frac{(k+1)^{k+1}}{k^k} = ... \to 1$. Actually ratio→1. My initial thought was wrong. Correct analysis shows divergence via asymptotic equivalence to harmonic. This traps students who misapply Ratio Test. Emphasizes simplification before testing.