← All Units← 返回单元列表 ← Course Hub← 课程主页
A B  C A L C U L U S
Chapter 10 · Solutions第10章 · 解析

Sequences & Series · Solutions数列与级数 · 解析

Worked Solutions to the AP-Style Practice SetAP 风格练习题配套解析

EASY MEDIUM HARD

Topics 10.1 - 10.15专题 10.1 至 10.15BC



PART ITopics 10.1 - 10.15专题 10.1 至 10.15

Multiple Choice Solutions选择题解析

Q1EASY 10.1 Partial Sums (Telescoping)10.1 部分和(裂项相消)No Calculator[2 marks]

$\displaystyle\sum_{n=1}^{\infty}\frac{1}{(2n-1)(2n+1)}=$$\displaystyle\sum_{n=1}^{\infty}\frac{1}{(2n-1)(2n+1)}=$

Answer:答案: (B) $\dfrac{1}{2}$

Split into partial fractions and telescope部分分式分解后裂项相消 M1·A1

Decompose the general term: $\dfrac{1}{(2n-1)(2n+1)}=\dfrac{1}{2}\left(\dfrac{1}{2n-1}-\dfrac{1}{2n+1}\right)$. (M1)分解通项:$\dfrac{1}{(2n-1)(2n+1)}=\dfrac{1}{2}\left(\dfrac{1}{2n-1}-\dfrac{1}{2n+1}\right)$。(M1)

$$ S_N=\frac{1}{2}\sum_{n=1}^{N}\left(\frac{1}{2n-1}-\frac{1}{2n+1}\right)=\frac{1}{2}\left(1-\frac{1}{2N+1}\right)\longrightarrow\frac{1}{2}. $$

(A1)

Insight.要点。 Only a telescoping series lets you find the exact sum without a closed formula: the partial fraction split turns every interior term into a canceling pair, leaving only the first surviving piece and a vanishing tail. Always write out $S_1, S_2, S_3$ by hand once to confirm which pieces actually cancel before trusting the pattern.只有裂项相消级数才能在没有闭式公式的情况下求出精确和:部分分式分解使每个内部项都能与相邻项配对相消,只留下首个未被抵消的部分与一个趋于零的尾项。在信任这一规律之前,务必先手写出 $S_1, S_2, S_3$ 以确认哪些项真正相消。
Q2EASY 10.2 Geometric Series10.2 等比级数No Calculator[2 marks]

$\displaystyle\sum_{n=1}^{\infty}\frac{5}{4^{n}}=$$\displaystyle\sum_{n=1}^{\infty}\frac{5}{4^{n}}=$

Answer:答案: (A) $\dfrac{5}{3}$

Identify $a$ and $r$, then apply $\frac{a}{1-r}$确定 $a$ 与 $r$,再代入 $\frac{a}{1-r}$ M1·A1

The series starts at $n=1$ with first term $a=5/4$ and common ratio $r=1/4$; since $|r|<1$, it converges. (M1)级数从 $n=1$ 开始,首项 $a=5/4$,公比 $r=1/4$;因 $|r|<1$,级数收敛。(M1)

$$ S=\frac{a}{1-r}=\frac{5/4}{1-1/4}=\frac{5/4}{3/4}=\frac{5}{3}. $$

(A1)

Insight.要点。 The most common slip on geometric-series questions is misreading the starting index: $a$ is the value of the first term actually summed, not the coefficient sitting in front of $r^n$. Here $a$ is the $n=1$ term $5/4$, not $5$.等比级数题最常见的失分点是读错起始指标:$a$ 应是实际求和的第一项的值,而非写在 $r^n$ 前面的系数。此处 $a$ 是 $n=1$ 时的项 $5/4$,而非 $5$。
Q3EASY 10.2 Geometric Series (Divergence)10.2 等比级数(发散)No Calculator[2 marks]

Which of the following series diverges?下列哪个级数发散?

Answer:答案: (C) $\displaystyle\sum\left(\dfrac{3}{2}\right)^{n}$

Check $|r|$ for each option检验各选项的 $|r|$ M1·A1

The ratios are $r=0.9,\ -0.5,\ 3/2,\ 1/3$ respectively; a geometric series converges precisely when $|r|<1$. (M1)各公比依次为 $r=0.9,\ -0.5,\ 3/2,\ 1/3$;等比级数当且仅当 $|r|<1$ 时收敛。(M1)

Only option (C) has $|r|=3/2>1$, so it is the only one that diverges. (A1)仅选项 (C) 满足 $|r|=3/2>1$,故只有它发散。(A1)

Insight.要点。 The convergence test for a geometric series is purely about the size of $r$, the sign never matters for convergence, only for the sign of the terms. A negative ratio like $-0.5$ still converges (it just alternates while shrinking); it is magnitude $\ge 1$ that kills convergence.等比级数的收敛判别只取决于 $r$ 的大小,符号对收敛性没有影响,只影响各项的正负。像 $-0.5$ 这样的负公比仍然收敛(只是在缩小的同时正负交替);真正破坏收敛性的是绝对值 $\ge 1$。
Q4EASY 10.3 nth Term Test10.3 第 n 项判别法No Calculator[2 marks]

$\displaystyle\sum_{n=1}^{\infty}\frac{2n^{2}+1}{3n^{2}+n}$$\displaystyle\sum_{n=1}^{\infty}\frac{2n^{2}+1}{3n^{2}+n}$

Answer:答案: (C) diverges发散

Take the limit of the general term求通项的极限 M1·A1

Divide numerator and denominator by $n^{2}$: $\displaystyle\lim_{n\to\infty}\frac{2n^{2}+1}{3n^{2}+n}=\lim_{n\to\infty}\frac{2+1/n^{2}}{3+1/n}=\frac{2}{3}$. (M1)分子分母同除以 $n^{2}$:$\displaystyle\lim_{n\to\infty}\frac{2n^{2}+1}{3n^{2}+n}=\lim_{n\to\infty}\frac{2+1/n^{2}}{3+1/n}=\frac{2}{3}$。(M1)

Since $\tfrac{2}{3}\ne 0$, the nth Term Test gives divergence immediately. (A1)因 $\tfrac{2}{3}\ne 0$,由第 n 项判别法立即得知级数发散。(A1)

Insight.要点。 This should always be the first test you try: computing $\lim a_n$ costs one line, and if it's nonzero you're finished. Option (A), "converges to $2/3$," is the classic trap of confusing the limit of the terms with the sum of the series, a series never converges to the limit of its terms.这应始终是你首先尝试的判别法:计算 $\lim a_n$ 只需一行,若非零则问题立即解决。选项 (A)"收敛到 $2/3$"是把项的极限误当作级数和的经典陷阱,级数绝不会收敛到其通项的极限。
Q5MEDIUM 10.4 Integral Test10.4 积分判别法No Calculator[2 marks]

$\displaystyle\sum_{n=2}^{\infty}\frac{1}{n\ln n}$$\displaystyle\sum_{n=2}^{\infty}\frac{1}{n\ln n}$

Answer:答案: (B) diverges by the integral test由积分判别法发散

Verify conditions, then evaluate the improper integral验证条件,再计算反常积分 M1·A1

Let $f(x)=\dfrac{1}{x\ln x}$: it is positive, continuous, and decreasing for $x\ge 2$, so the integral test applies. (M1)设 $f(x)=\dfrac{1}{x\ln x}$:在 $x\ge 2$ 上为正、连续且单调递减,故可用积分判别法。(M1)

$$ \int_2^{\infty}\frac{dx}{x\ln x}=\lim_{b\to\infty}\bigl[\ln(\ln x)\bigr]_2^{b}=\lim_{b\to\infty}\ln(\ln b)-\ln(\ln 2)=\infty. $$

The improper integral diverges, so the series diverges by the integral test. (A1)该反常积分发散,故由积分判别法,级数发散。(A1)

Insight.要点。 This series is a favorite trap because $\tfrac{1}{n\ln n}\to 0$ even more slowly than $\tfrac{1}{n}$, so it feels like it "should" converge; the nth term test is powerless here since the limit is $0$. Only the integral test (via the substitution $u=\ln x$) exposes the divergence, since $\ln(\ln x)\to\infty$, just extremely slowly.该级数是一个经典陷阱,因为 $\tfrac{1}{n\ln n}\to 0$ 的速度甚至比 $\tfrac{1}{n}$ 还慢,让人误以为它"应该"收敛;此处第 n 项判别法无能为力,因为极限为 $0$。只有积分判别法(通过代换 $u=\ln x$)才能揭示其发散性,因为 $\ln(\ln x)\to\infty$,只是速度极其缓慢。
Q6EASY 10.5 $p$-Series10.5 $p$ 级数No Calculator[2 marks]

Which series converges?下列哪个级数收敛?

Answer:答案: (C) $\displaystyle\sum\frac{1}{n^{1.5}}$

Read off $p$ for each option读出各选项的 $p$ 值 M1·A1

The exponents are $p=1,\ 0.5,\ 1.5,\ 0.9$ respectively; a $p$-series converges exactly when $p>1$. (M1)各指数依次为 $p=1,\ 0.5,\ 1.5,\ 0.9$;$p$ 级数当且仅当 $p>1$ 时收敛。(M1)

Only $p=1.5>1$ in option (C) satisfies this, so it is the only convergent series. (A1)仅选项 (C) 的 $p=1.5>1$ 满足此条件,故只有它收敛。(A1)

Insight.要点。 $p$-series classification is instant recognition, not computation: read the exponent, compare to $1$, done. The boundary $p=1$ (option A, the harmonic series) is the case students most often misjudge as convergent since $1/n\to 0$; it is the sharpest illustration that "terms shrink to zero" is not enough.$p$ 级数的判别应做到一眼识别,而非计算:读出指数,与 $1$ 比较即可。边界情形 $p=1$(选项 A,即调和级数)是学生最容易误判为收敛的情形,因为 $1/n\to 0$;这正是"各项趋于零"并不足够的最典型例证。
Q7MEDIUM 10.5 Harmonic Series10.5 调和级数No Calculator[2 marks]

Which of the following is a true statement about the harmonic series $\displaystyle\sum_{n=1}^{\infty}\frac{1}{n}$?下列关于调和级数 $\displaystyle\sum_{n=1}^{\infty}\frac{1}{n}$ 的说法中,哪个正确?

Answer:答案: (C)

Eliminate the distractors one by one逐一排除干扰项 M1·A1

(A) is false: it is the alternating harmonic series $\sum(-1)^{n+1}/n$ that converges to $\ln 2$, not $\sum 1/n$ itself. (B) is false: $\int_1^{\infty}\tfrac{1}{x}\,dx=\lim_{b\to\infty}\ln b=\infty$, so the integral test shows divergence, not convergence. (D) is false as a justification: since $\lim 1/n=0$, the nth term test is inconclusive here, not a valid reason for divergence. (M1)(A) 错误:收敛到 $\ln 2$ 的是交错调和级数 $\sum(-1)^{n+1}/n$,而非 $\sum 1/n$ 本身。(B) 错误:$\int_1^{\infty}\tfrac{1}{x}\,dx=\lim_{b\to\infty}\ln b=\infty$,积分判别法显示的是发散而非收敛。(D) 作为理由不成立:因 $\lim 1/n=0$,第 n 项判别法在此不能下结论,不能作为发散的正确依据。(M1)

(C) is the correct statement: the series does diverge (by the integral test above), even though the individual terms shrink to $0$. (A1)(C) 为正确说法:级数确实发散(由上面的积分判别法可知),尽管各项本身趋于 $0$。(A1)

Insight.要点。 The harmonic series is the single most important counterexample in this unit: it is the standard proof that $\lim a_n=0$ never guarantees convergence. Keep it and the alternating harmonic series (which does converge, to $\ln 2$) mentally paired, since exam distractors love to swap their properties.调和级数是本单元中最重要的反例:它是"$\lim a_n=0$ 绝不保证收敛"这一结论的标准证明。应将它与交错调和级数(后者确实收敛,到 $\ln 2$)在脑中配对记忆,因为考试干扰项最爱互换二者的性质。
Q8EASY 10.6 Direct Comparison Test10.6 比较判别法No Calculator[2 marks]

Since $0\le\dfrac{1}{n^{3}+1}\le\dfrac{1}{n^{3}}$ for all $n\ge 1$, and $\displaystyle\sum\frac{1}{n^{3}}$ converges, what does the Direct Comparison Test conclude about $\displaystyle\sum_{n=1}^{\infty}\frac{1}{n^{3}+1}$?因对所有 $n\ge 1$ 有 $0\le\dfrac{1}{n^{3}+1}\le\dfrac{1}{n^{3}}$,且 $\displaystyle\sum\frac{1}{n^{3}}$ 收敛,比较判别法对 $\displaystyle\sum_{n=1}^{\infty}\frac{1}{n^{3}+1}$ 得出什么结论?

Answer:答案: (A) converges收敛

Identify the bounding series and its behavior确定用于比较的级数及其收敛性 M1·A1

$\displaystyle\sum\frac{1}{n^{3}}$ is a $p$-series with $p=3>1$, so it converges. (M1)$\displaystyle\sum\frac{1}{n^{3}}$ 是 $p$ 级数,$p=3>1$,故收敛。(M1)

Since the given series is squeezed between $0$ and a convergent series termwise, the Direct Comparison Test concludes $\displaystyle\sum\frac{1}{n^{3}+1}$ also converges. (A1)因给定级数逐项夹在 $0$ 与一个收敛级数之间,由比较判别法知 $\displaystyle\sum\frac{1}{n^{3}+1}$ 同样收敛。(A1)

Insight.要点。 DCT direction is easy to misremember: a series smaller than a convergent series converges; a series larger than a divergent series diverges. The other two combinations, smaller than divergent, larger than convergent, tell you nothing, which is exactly why the LCT exists for cases where the inequality itself is inconvenient to prove.比较判别法的方向容易记混:小于一个收敛级数则收敛;大于一个发散级数则发散。另外两种组合(小于发散级数、大于收敛级数)则无法得出任何结论,这正是当不等式本身难以直接证明时,需要用极限比较判别法的原因。
Q9MEDIUM 10.6 Limit Comparison Test10.6 极限比较判别法No Calculator[2 marks]

$\displaystyle\sum_{n=1}^{\infty}\frac{n+2}{n^{3}-n+1}$$\displaystyle\sum_{n=1}^{\infty}\frac{n+2}{n^{3}-n+1}$

Answer:答案: (A) converges (LCT with $1/n^{2}$)收敛(用 $1/n^{2}$ 作极限比较)

Identify the dominant behavior, then compute the limit ratio确定主导行为,再计算极限比值 M1·A1

For large $n$, $a_n=\dfrac{n+2}{n^{3}-n+1}\approx\dfrac{n}{n^{3}}=\dfrac{1}{n^{2}}$, so compare with $b_n=1/n^{2}$ (a convergent $p$-series, $p=2$). (M1)当 $n$ 较大时,$a_n=\dfrac{n+2}{n^{3}-n+1}\approx\dfrac{n}{n^{3}}=\dfrac{1}{n^{2}}$,故与 $b_n=1/n^{2}$(收敛的 $p$ 级数,$p=2$)比较。(M1)

$$ \lim_{n\to\infty}\frac{a_n}{b_n}=\lim_{n\to\infty}\frac{n+2}{n^{3}-n+1}\cdot n^{2}=\lim_{n\to\infty}\frac{n^{3}+2n^{2}}{n^{3}-n+1}=1. $$

Since $0<1<\infty$ and $\sum 1/n^{2}$ converges, the LCT gives that $\sum a_n$ converges too. (A1)因 $0<1<\infty$ 且 $\sum 1/n^{2}$ 收敛,由极限比较判别法知 $\sum a_n$ 也收敛。(A1)

Insight.要点。 Picking $b_n$ is a matter of dividing the leading term of the numerator by the leading term of the denominator, ignore every lower-order piece. The LCT is preferred over DCT here precisely because $\dfrac{n+2}{n^{3}-n+1}\le\dfrac{1}{n^{2}}$ would be annoying to verify directly, while the limit computation is routine algebra.选取 $b_n$ 的方法是用分子的最高次项除以分母的最高次项,忽略所有低阶项。此处优先选用极限比较判别法而非直接比较判别法,正是因为直接验证 $\dfrac{n+2}{n^{3}-n+1}\le\dfrac{1}{n^{2}}$ 较为麻烦,而极限计算只是常规代数运算。
Q10MEDIUM 10.7 Alternating Series Test10.7 交错级数判别法No Calculator[2 marks]

$\displaystyle\sum_{n=1}^{\infty}(-1)^{n}\frac{n}{n^{2}+1}$$\displaystyle\sum_{n=1}^{\infty}(-1)^{n}\frac{n}{n^{2}+1}$

Answer:答案: (A) converges by the Alternating Series Test由交错级数判别法收敛

Verify all three AST conditions on $b_n=n/(n^{2}+1)$对 $b_n=n/(n^{2}+1)$ 验证交错级数判别法的三个条件 M1·M1·A1

$b_n=\dfrac{n}{n^{2}+1}>0$ for all $n\ge 1$. (M1)对所有 $n\ge 1$,$b_n=\dfrac{n}{n^{2}+1}>0$。(M1)

Letting $f(x)=\dfrac{x}{x^{2}+1}$, $f'(x)=\dfrac{(x^{2}+1)-x(2x)}{(x^{2}+1)^{2}}=\dfrac{1-x^{2}}{(x^{2}+1)^{2}}<0$ for $x>1$, so $b_n$ is eventually decreasing; also $\displaystyle\lim_{n\to\infty}b_n=0$. (M1)令 $f(x)=\dfrac{x}{x^{2}+1}$,则 $f'(x)=\dfrac{(x^{2}+1)-x(2x)}{(x^{2}+1)^{2}}=\dfrac{1-x^{2}}{(x^{2}+1)^{2}}<0$($x>1$ 时),故 $b_n$ 最终单调递减;且 $\displaystyle\lim_{n\to\infty}b_n=0$。(M1)

All three conditions hold, so the series converges by the Alternating Series Test. (A1)三个条件均满足,故由交错级数判别法,级数收敛。(A1)

Insight.要点。 Option (C) is a trap: $\displaystyle\sum\frac{n}{n^{2}+1}$ (the series of absolute values) itself diverges by the LCT against $1/n$, so this series is only conditionally, not absolutely, convergent. The AST guarantees convergence but says nothing about which kind, that requires a separate check, which is exactly the subject of Topic 10.9.选项 (C) 是一个陷阱:绝对值级数 $\displaystyle\sum\frac{n}{n^{2}+1}$ 本身用 $1/n$ 作极限比较即可判定发散,故此级数只是条件收敛,而非绝对收敛。交错级数判别法只保证收敛,但不说明是哪种收敛,这需要另行检验,正是 10.9 节的主题。
Q11MEDIUM 10.8 Ratio Test10.8 比值判别法No Calculator[2 marks]

$\displaystyle\sum_{n=1}^{\infty}\frac{2^{n}}{n!}$$\displaystyle\sum_{n=1}^{\infty}\frac{2^{n}}{n!}$

Answer:答案: (A) converges absolutely (ratio test)绝对收敛(比值判别法)

Compute $L=\lim|a_{n+1}/a_n|$计算 $L=\lim|a_{n+1}/a_n|$ M1·A1

$\left|\dfrac{a_{n+1}}{a_n}\right|=\dfrac{2^{n+1}/(n+1)!}{2^{n}/n!}=\dfrac{2\cdot n!}{(n+1)!}=\dfrac{2}{n+1}$. (M1)$\left|\dfrac{a_{n+1}}{a_n}\right|=\dfrac{2^{n+1}/(n+1)!}{2^{n}/n!}=\dfrac{2\cdot n!}{(n+1)!}=\dfrac{2}{n+1}$。(M1)

$L=\displaystyle\lim_{n\to\infty}\frac{2}{n+1}=0<1$, so by the ratio test the series converges absolutely (its terms are already positive, so absolute convergence is just convergence here). (A1)$L=\displaystyle\lim_{n\to\infty}\frac{2}{n+1}=0<1$,故由比值判别法,级数绝对收敛(因各项已为正,此处绝对收敛即为收敛)。(A1)

Insight.要点。 Any series with an $n!$ in it should trigger the ratio test on sight: the factorial ratio $\dfrac{n!}{(n+1)!}=\dfrac{1}{n+1}$ collapses almost everything else in the expression, which is exactly why factorials and exponentials pair so cleanly with this test.凡是出现 $n!$ 的级数,应立即想到比值判别法:阶乘之比 $\dfrac{n!}{(n+1)!}=\dfrac{1}{n+1}$ 几乎能化简表达式中的其余所有部分,这正是阶乘与指数如此适合该判别法的原因。
Q12MEDIUM 10.9 Absolute vs. Conditional10.9 绝对收敛与条件收敛No Calculator[2 marks]

$\displaystyle\sum_{n=1}^{\infty}\frac{(-1)^{n+1}}{\sqrt{n}}$ is$\displaystyle\sum_{n=1}^{\infty}\frac{(-1)^{n+1}}{\sqrt{n}}$ 是

Answer:答案: (B) conditionally convergent条件收敛

Test $\sum|a_n|$ first, then the original series先检验 $\sum|a_n|$,再检验原级数 M1·A1

$\displaystyle\sum\left|\frac{(-1)^{n+1}}{\sqrt{n}}\right|=\sum\frac{1}{n^{1/2}}$ is a $p$-series with $p=1/2\le 1$, so it diverges: the series is not absolutely convergent. (M1)$\displaystyle\sum\left|\frac{(-1)^{n+1}}{\sqrt{n}}\right|=\sum\frac{1}{n^{1/2}}$ 是 $p$ 级数,$p=1/2\le 1$,故发散:该级数不是绝对收敛。(M1)

But $b_n=1/\sqrt{n}$ is positive, decreasing, and $\to 0$, so the original series converges by the AST. Converging overall while $\sum|a_n|$ diverges is exactly conditional convergence. (A1)但 $b_n=1/\sqrt{n}$ 为正、递减且 $\to 0$,故由交错级数判别法,原级数收敛。整体收敛而 $\sum|a_n|$ 发散,正是条件收敛的定义。(A1)

Insight.要点。 The three-way classification is always a two-step process: test $\sum|a_n|$ first (any convergence test is fair game there, since it's positive-term); only if that diverges do you go back and test $\sum a_n$ itself. Skipping straight to "it converges by AST" without checking $\sum|a_n|$ throws away half the required information.绝对收敛、条件收敛、发散这三分类的判断总是两步走:先检验 $\sum|a_n|$(此时可用任何判别法,因为它是正项级数);只有当其发散时,才回头检验 $\sum a_n$ 本身。若跳过检验 $\sum|a_n|$ 而直接下结论"由交错级数判别法收敛",就遗漏了所需信息的一半。
Q13HARD 10.10 Alternating Series Error Bound10.10 交错级数误差界Calculator[2 marks]

For $\displaystyle\sum_{n=1}^{\infty}\frac{(-1)^{n+1}}{n!}$, what is the smallest $n$ such that the partial sum $S_n$ approximates the sum with error less than $0.001$?对 $\displaystyle\sum_{n=1}^{\infty}\frac{(-1)^{n+1}}{n!}$,使部分和 $S_n$ 的误差小于 $0.001$ 的最小 $n$ 是多少?

Answer:答案: (C) $6$

Bound the error by the first omitted term用第一个被舍弃项界定误差 M1·A1

The error bound requires $|a_{n+1}|=\dfrac{1}{(n+1)!}<0.001$, i.e. $(n+1)!>1000$. (M1)误差界要求 $|a_{n+1}|=\dfrac{1}{(n+1)!}<0.001$,即 $(n+1)!>1000$。(M1)

Check factorials: $6!=720<1000$, but $7!=5040>1000$. So the smallest $n+1$ that works is $7$, giving $n=6$. (A1)检验阶乘:$6!=720<1000$,而 $7!=5040>1000$。故满足条件的最小 $n+1$ 为 $7$,即 $n=6$。(A1)

Insight.要点。 The error bound always talks about the term after the last one you kept: if you stop at $S_n$, the bounding term is $a_{n+1}$, not $a_n$. This off-by-one is the single most common slip on this topic, always write out which index the sum actually stops at before applying the inequality.误差界谈论的始终是你保留的最后一项之后的那一项:若在 $S_n$ 处停止,用于界定误差的项是 $a_{n+1}$,而非 $a_n$。这种错位一是本主题最常见的失误,应用不等式前务必先写清求和实际停在哪个指标。
Q14EASY 10.11 Taylor Polynomials10.11 泰勒多项式No Calculator[2 marks]

In the 4th-degree Maclaurin polynomial for $f(x)=\cos x$, the coefficient of $x^{4}$ is在 $f(x)=\cos x$ 的 4 次麦克劳林多项式中,$x^{4}$ 的系数为

Answer:答案: (A) $\dfrac{1}{24}$

Apply the coefficient formula $f^{(k)}(0)/k!$代入系数公式 $f^{(k)}(0)/k!$ M1·A1

$f(x)=\cos x$, $f''''(x)=\cos x$ (the 4th derivative returns to $\cos x$), so $f''''(0)=\cos 0=1$. (M1)$f(x)=\cos x$,$f''''(x)=\cos x$(四阶导数回到 $\cos x$),故 $f''''(0)=\cos 0=1$。(M1)

The coefficient of $x^{4}$ in $P_4(x)$ is $\dfrac{f''''(0)}{4!}=\dfrac{1}{24}$. (A1)$P_4(x)$ 中 $x^{4}$ 的系数为 $\dfrac{f''''(0)}{4!}=\dfrac{1}{24}$。(A1)

Insight.要点。 This matches the memorized series $\cos x=1-\tfrac{x^{2}}{2!}+\tfrac{x^{4}}{4!}-\cdots$ term by term: the $x^{4}$ coefficient is $+\tfrac{1}{4!}$, positive, since the alternating sign pattern is $+,-,+,-$ starting at the constant term, so the $x^4$ term (the third term) carries a $+$.这与必记级数 $\cos x=1-\tfrac{x^{2}}{2!}+\tfrac{x^{4}}{4!}-\cdots$ 逐项吻合:$x^{4}$ 的系数为 $+\tfrac{1}{4!}$,为正,因为从常数项开始正负交替模式为 $+,-,+,-$,故 $x^4$ 项(第三项)取正号。
Q15HARD 10.12 Lagrange Error Bound10.12 拉格朗日误差界Calculator[2 marks]

Let $P_3(x)$ be the 3rd-degree Taylor polynomial for $f(x)=e^{x}$ centered at $a=0$. If $|f^{(4)}(c)|\le 3$ for $0\le c\le 1$, the Lagrange error bound for $|f(1)-P_3(1)|$ is设 $P_3(x)$ 为 $f(x)=e^{x}$ 在 $a=0$ 处的 3 次泰勒多项式。若对 $0\le c\le 1$ 有 $|f^{(4)}(c)|\le 3$,则 $|f(1)-P_3(1)|$ 的拉格朗日误差界为

Answer:答案: (B) $\dfrac{1}{8}$

Substitute $M$, $n$, and $|x-a|$ into the Lagrange bound将 $M$、$n$、$|x-a|$ 代入拉格朗日误差界 M1·A1

Here $M=3$, $n=3$ (so $n+1=4$), and $|x-a|=|1-0|=1$. (M1)此处 $M=3$,$n=3$(故 $n+1=4$),$|x-a|=|1-0|=1$。(M1)

$$ |R_3(1)|\le\frac{M}{(n+1)!}|x-a|^{n+1}=\frac{3}{4!}\cdot 1^{4}=\frac{3}{24}=\frac{1}{8}. $$

(A1)

Insight.要点。 The Lagrange bound uses $(n+1)!$, not $n!$, and the exponent on $|x-a|$ is also $n+1$, both indices shift up by one from the degree of the polynomial, since the bound involves the next derivative beyond the ones used in $P_n$. Forgetting this shift and using $n!$ or exponent $n$ is the most common error on this topic.拉格朗日误差界使用的是 $(n+1)!$ 而非 $n!$,$|x-a|$ 的指数同样是 $n+1$,两个指标都比多项式的次数多 $1$,因为该误差界涉及的是 $P_n$ 所用导数之后的那一阶导数。忘记这一偏移、误用 $n!$ 或指数 $n$,是此主题最常见的错误。
Q16MEDIUM 10.13 Radius of Convergence10.13 收敛半径No Calculator[2 marks]

The radius of convergence of $\displaystyle\sum_{n=1}^{\infty}\frac{(x+1)^{n}}{n\cdot 3^{n}}$ is$\displaystyle\sum_{n=1}^{\infty}\frac{(x+1)^{n}}{n\cdot 3^{n}}$ 的收敛半径为

Answer:答案: (C) $3$

Apply the ratio test directly to the power series直接对幂级数应用比值判别法 M1·A1

$\left|\dfrac{a_{n+1}}{a_n}\right|=\left|\dfrac{(x+1)^{n+1}}{(n+1)3^{n+1}}\cdot\dfrac{n\cdot 3^{n}}{(x+1)^{n}}\right|=\dfrac{|x+1|}{3}\cdot\dfrac{n}{n+1}\longrightarrow\dfrac{|x+1|}{3}$. (M1)$\left|\dfrac{a_{n+1}}{a_n}\right|=\left|\dfrac{(x+1)^{n+1}}{(n+1)3^{n+1}}\cdot\dfrac{n\cdot 3^{n}}{(x+1)^{n}}\right|=\dfrac{|x+1|}{3}\cdot\dfrac{n}{n+1}\longrightarrow\dfrac{|x+1|}{3}$。(M1)

The series converges when $\dfrac{|x+1|}{3}<1$, i.e. $|x+1|<3$, so $R=3$. (A1)当 $\dfrac{|x+1|}{3}<1$ 即 $|x+1|<3$ 时级数收敛,故 $R=3$。(A1)

Insight.要点。 The radius of convergence is read directly off the number that $|x-c|$ gets compared to once the ratio test limit simplifies, here $3$, regardless of what extra polynomial factor in $n$ (like $n/(n+1)$) is riding along, since that factor always $\to 1$. Do not confuse the $3$ in the denominator with the center of the series, which is $c=-1$, read from $(x+1)^n=(x-(-1))^n$.收敛半径就是比值判别法极限化简后,$|x-c|$ 所对比的那个数,此处为 $3$,与其中携带的关于 $n$ 的多项式因子(如 $n/(n+1)$)无关,因为该因子总是趋于 $1$。不要将分母中的 $3$ 与级数的中心混淆,中心应为 $c=-1$,由 $(x+1)^n=(x-(-1))^n$ 读出。
Q17MEDIUM 10.14 Maclaurin Series10.14 麦克劳林级数No Calculator[2 marks]

Which series represents $\cos(x^{2})$?下列哪个级数表示 $\cos(x^{2})$?

Answer:答案: (A) $1-\dfrac{x^{4}}{2!}+\dfrac{x^{8}}{4!}-\cdots$

Substitute $u=x^{2}$ into the known series for $\cos u$将 $u=x^{2}$ 代入已知的 $\cos u$ 级数 M1·A1

Start from $\cos u=\displaystyle\sum_{n=0}^{\infty}\frac{(-1)^{n}u^{2n}}{(2n)!}=1-\frac{u^{2}}{2!}+\frac{u^{4}}{4!}-\cdots$, and let $u=x^{2}$. (M1)从 $\cos u=\displaystyle\sum_{n=0}^{\infty}\frac{(-1)^{n}u^{2n}}{(2n)!}=1-\frac{u^{2}}{2!}+\frac{u^{4}}{4!}-\cdots$ 出发,令 $u=x^{2}$。(M1)

$$ \cos(x^{2})=1-\frac{(x^{2})^{2}}{2!}+\frac{(x^{2})^{4}}{4!}-\cdots=1-\frac{x^{4}}{2!}+\frac{x^{8}}{4!}-\cdots, $$

which is option (A). (A1)即选项 (A)。(A1)

Insight.要点。 Substitution never changes the sign pattern or the factorial denominators, it only changes what power of $x$ appears: $\cos u$'s $u^{2n}$ becomes $(x^{2})^{2n}=x^{4n}$. Option (B) is the trap for students who forget to substitute at all (that's just $\cos x$), and option (C) confuses $\cos$'s even powers with $\sin$'s odd ones.代换不会改变符号规律或阶乘分母,只会改变 $x$ 的幂次:$\cos u$ 中的 $u^{2n}$ 变为 $(x^{2})^{2n}=x^{4n}$。选项 (B) 是给忘记代换的学生设的陷阱(那只是 $\cos x$ 本身),选项 (C) 则混淆了 $\cos$ 的偶次幂与 $\sin$ 的奇次幂。
Q18HARD 10.15 Representing Functions as Series10.15 幂级数表示函数No Calculator[2 marks]

A power series representation of $\dfrac{x}{1-x^{3}}$ is$\dfrac{x}{1-x^{3}}$ 的幂级数表示为

Answer:答案: (A) $\displaystyle\sum_{n=0}^{\infty}x^{3n+1}$

Substitute, then multiply by $x$先代换,再乘以 $x$ M1·A1

Start from $\dfrac{1}{1-u}=\displaystyle\sum_{n=0}^{\infty}u^{n}$ for $|u|<1$, and let $u=x^{3}$: $\dfrac{1}{1-x^{3}}=\displaystyle\sum_{n=0}^{\infty}x^{3n}$. (M1)从 $\dfrac{1}{1-u}=\displaystyle\sum_{n=0}^{\infty}u^{n}$($|u|<1$)出发,令 $u=x^{3}$:$\dfrac{1}{1-x^{3}}=\displaystyle\sum_{n=0}^{\infty}x^{3n}$。(M1)

Multiply every term by $x$: $\dfrac{x}{1-x^{3}}=\displaystyle\sum_{n=0}^{\infty}x\cdot x^{3n}=\displaystyle\sum_{n=0}^{\infty}x^{3n+1}=x+x^{4}+x^{7}+\cdots$. (A1)每项乘以 $x$:$\dfrac{x}{1-x^{3}}=\displaystyle\sum_{n=0}^{\infty}x\cdot x^{3n}=\displaystyle\sum_{n=0}^{\infty}x^{3n+1}=x+x^{4}+x^{7}+\cdots$。(A1)

Insight.要点。 Building a new power series is almost always substitute-then-multiply (or substitute-then-integrate/differentiate): identify the geometric skeleton $\tfrac{1}{1-u}$ hiding inside the expression first, substitute for $u$, and only then apply whatever outer operation (here, multiplying by $x$) the original function needed.构造新的幂级数几乎总是"先代换、再乘(或再积分/求导)":先识别表达式中隐藏的等比级数骨架 $\tfrac{1}{1-u}$,代换 $u$,然后再应用原函数所需的外层运算(此处为乘以 $x$)。
PART IIShow All Work展示完整解题过程

Free-Response Solutions自由解答题解析

FRQ 1EASY 10.1 / 10.3 / 10.8 Classify Convergence10.1 / 10.3 / 10.8 判别收敛性No Calculator[6 marks]

For each series below, state whether it converges or diverges, name the test used, and (if it converges by a direct computation) give the sum: (a) $\displaystyle\sum_{n=1}^{\infty}\frac{n}{n+5}$; (b) $\displaystyle\sum_{n=1}^{\infty}\frac{1}{n(n+2)}$; (c) $\displaystyle\sum_{n=0}^{\infty}\frac{3^{n}}{n!}$.对下列各级数,判断其是收敛还是发散,写出所用判别法,并(若可直接求和)给出和:(a) $\displaystyle\sum_{n=1}^{\infty}\frac{n}{n+5}$;(b) $\displaystyle\sum_{n=1}^{\infty}\frac{1}{n(n+2)}$;(c) $\displaystyle\sum_{n=0}^{\infty}\frac{3^{n}}{n!}$。

Answers:答案:  (a) diverges发散  ·  (b) converges to收敛到 $\dfrac{3}{4}$  ·  (c) converges收敛

(a) nth Term Test(a) 第 n 项判别法 M1·A1

$\displaystyle\lim_{n\to\infty}\frac{n}{n+5}=1\ne 0$. (M1)$\displaystyle\lim_{n\to\infty}\frac{n}{n+5}=1\ne 0$。(M1)

By the nth Term Test, the series diverges. (A1)由第 n 项判别法,级数发散。(A1)

(b) Telescoping (partial fractions)(b) 裂项相消(部分分式) M1·A1

$\dfrac{1}{n(n+2)}=\dfrac{1}{2}\left(\dfrac{1}{n}-\dfrac{1}{n+2}\right)$, so $S_N=\dfrac{1}{2}\left(1+\dfrac{1}{2}-\dfrac{1}{N+1}-\dfrac{1}{N+2}\right)$. (M1)$\dfrac{1}{n(n+2)}=\dfrac{1}{2}\left(\dfrac{1}{n}-\dfrac{1}{n+2}\right)$,故 $S_N=\dfrac{1}{2}\left(1+\dfrac{1}{2}-\dfrac{1}{N+1}-\dfrac{1}{N+2}\right)$。(M1)

As $N\to\infty$, the sum $\to\dfrac{1}{2}\left(\dfrac{3}{2}\right)=\dfrac{3}{4}$. (A1)当 $N\to\infty$ 时,和 $\to\dfrac{1}{2}\left(\dfrac{3}{2}\right)=\dfrac{3}{4}$。(A1)

(c) Ratio Test(c) 比值判别法 M1·A1

$\left|\dfrac{a_{n+1}}{a_n}\right|=\dfrac{3}{n+1}\to 0<1$. (M1)$\left|\dfrac{a_{n+1}}{a_n}\right|=\dfrac{3}{n+1}\to 0<1$。(M1)

By the ratio test, the series converges (in fact, to $e^{3}$, though the sum is not required here). (A1)由比值判别法,级数收敛(实际上收敛到 $e^{3}$,但此处不要求求和)。(A1)

Insight.要点。 These three cases fix the order of attack that Topic 10.16's flowchart formalizes: always try the nth Term Test first (a costs one line and sometimes finishes the problem), recognize telescoping/geometric structure next since those give an exact sum, and reach for the ratio test only when factorials or exponentials appear.这三个例子正是 10.16 节流程图所规范化的进攻顺序:始终先尝试第 n 项判别法(只需一行,有时就能直接解决问题),接着识别裂项相消/等比结构(因为它们能给出精确和),只有出现阶乘或指数时才使用比值判别法。
FRQ 2MEDIUM 10.7 / 10.9 / 10.10 Alternating Series Analysis10.7 / 10.9 / 10.10 交错级数综合分析No Calculator[9 marks]

Let $\displaystyle\sum_{n=1}^{\infty}\frac{(-1)^{n+1}}{2n+1}$.设 $\displaystyle\sum_{n=1}^{\infty}\frac{(-1)^{n+1}}{2n+1}$。

Answers:答案:  (a) converges by AST由交错级数判别法收敛  ·  (b) conditionally convergent条件收敛  ·  (c) $n=9$

(a) Verify all three AST conditions on $b_n=1/(2n+1)$(a) 对 $b_n=1/(2n+1)$ 验证三个条件 M1·M1·A1·R1

$b_n=\dfrac{1}{2n+1}>0$ for all $n\ge 1$. (M1)对所有 $n\ge 1$,$b_n=\dfrac{1}{2n+1}>0$。(M1)

$b_{n+1}=\dfrac{1}{2n+3}<\dfrac{1}{2n+1}=b_n$ since the denominator is strictly increasing, so $\{b_n\}$ is decreasing. (M1)因分母严格递增,$b_{n+1}=\dfrac{1}{2n+3}<\dfrac{1}{2n+1}=b_n$,故 $\{b_n\}$ 递减。(M1)

$\displaystyle\lim_{n\to\infty}b_n=\lim_{n\to\infty}\frac{1}{2n+1}=0$. (A1)$\displaystyle\lim_{n\to\infty}b_n=\lim_{n\to\infty}\frac{1}{2n+1}=0$。(A1)

All three conditions hold, so the series converges by the Alternating Series Test. (R1)三个条件均满足,故由交错级数判别法,级数收敛。(R1)

(b) Test $\sum|a_n|$ via the Limit Comparison Test(b) 用极限比较判别法检验 $\sum|a_n|$ M1·A1·R1

$\displaystyle\sum|a_n|=\sum\frac{1}{2n+1}$; compare with $b_n=1/n$: $\displaystyle\lim_{n\to\infty}\frac{1/(2n+1)}{1/n}=\lim_{n\to\infty}\frac{n}{2n+1}=\frac{1}{2}$. (M1)$\displaystyle\sum|a_n|=\sum\frac{1}{2n+1}$;与 $b_n=1/n$ 比较:$\displaystyle\lim_{n\to\infty}\frac{1/(2n+1)}{1/n}=\lim_{n\to\infty}\frac{n}{2n+1}=\frac{1}{2}$。(M1)

Since $0<\tfrac12<\infty$ and $\sum 1/n$ (harmonic) diverges, the LCT gives $\sum|a_n|$ diverges: the series is not absolutely convergent. (A1)因 $0<\tfrac12<\infty$ 且 $\sum 1/n$(调和级数)发散,由极限比较判别法知 $\sum|a_n|$ 发散:该级数不是绝对收敛。(A1)

Since the original series converges (part (a)) but $\sum|a_n|$ diverges, the series is conditionally convergent. (R1)因原级数收敛((a) 部分)而 $\sum|a_n|$ 发散,该级数为条件收敛。(R1)

(c) Solve the error-bound inequality(c) 求解误差界不等式 M1·A1

Need $|a_{n+1}|=\dfrac{1}{2(n+1)+1}=\dfrac{1}{2n+3}<0.05$, i.e. $2n+3>20$, so $n>8.5$. (M1)需要 $|a_{n+1}|=\dfrac{1}{2(n+1)+1}=\dfrac{1}{2n+3}<0.05$,即 $2n+3>20$,故 $n>8.5$。(M1)

The smallest integer satisfying this is $n=9$ (check: $\tfrac{1}{2(9)+3}=\tfrac{1}{21}\approx 0.048<0.05$, while $n=8$ gives $\tfrac{1}{19}\approx 0.053>0.05$). (A1)满足条件的最小整数为 $n=9$(检验:$\tfrac{1}{2(9)+3}=\tfrac{1}{21}\approx 0.048<0.05$,而 $n=8$ 时 $\tfrac{1}{19}\approx 0.053>0.05$)。(A1)

Insight.要点。 This question sequences all three ideas of the topic on purpose: AST only proves convergence (a), it never says what kind (b), and the error bound (c) only makes sense once convergence itself is established. Note in (b) that the LCT is run on the positive series $\sum|a_n|$, not on the original alternating series, mixing the two is a common and serious error.本题有意将本主题的三个思想串联在一起:交错级数判别法只能证明收敛 (a),它从不说明是哪种收敛 (b),而误差界 (c) 只有在收敛性本身确立之后才有意义。注意 (b) 中极限比较判别法是对正项级数 $\sum|a_n|$ 使用,而非对原交错级数使用,混淆二者是常见且严重的错误。
FRQ 3MEDIUM 10.11 / 10.12 Taylor Polynomial & Error10.11 / 10.12 泰勒多项式与误差Calculator[9 marks]

Let $f(x)=\ln x$, and let $P_3(x)$ be the 3rd-degree Taylor polynomial for $f$ centered at $a=1$.设 $f(x)=\ln x$,$P_3(x)$ 为 $f$ 在 $a=1$ 处的 3 次泰勒多项式。

Answers:答案:  (a) $P_3(x)=(x-1)-\dfrac{(x-1)^{2}}{2}+\dfrac{(x-1)^{3}}{3}$  ·  (b) $\ln(1.2)\approx 0.18267$  ·  (c) $|R_3|\le 0.0004$

(a) Compute derivatives at $a=1$(a) 计算在 $a=1$ 处的各阶导数 M1·M1·A1

$f(1)=0$; $f'(x)=1/x\Rightarrow f'(1)=1$; $f''(x)=-1/x^{2}\Rightarrow f''(1)=-1$; $f'''(x)=2/x^{3}\Rightarrow f'''(1)=2$. (M1)$f(1)=0$;$f'(x)=1/x\Rightarrow f'(1)=1$;$f''(x)=-1/x^{2}\Rightarrow f''(1)=-1$;$f'''(x)=2/x^{3}\Rightarrow f'''(1)=2$。(M1)

Substitute into $P_3(x)=f(1)+f'(1)(x-1)+\dfrac{f''(1)}{2!}(x-1)^{2}+\dfrac{f'''(1)}{3!}(x-1)^{3}$: (M1)代入 $P_3(x)=f(1)+f'(1)(x-1)+\dfrac{f''(1)}{2!}(x-1)^{2}+\dfrac{f'''(1)}{3!}(x-1)^{3}$:(M1)

$$ P_3(x)=(x-1)-\frac{(x-1)^{2}}{2}+\frac{(x-1)^{3}}{3}. $$

(A1)

(b) Substitute $x=1.2$(b) 代入 $x=1.2$ M1·A1

With $x-1=0.2$: (M1)取 $x-1=0.2$:(M1)

$$ P_3(1.2)=0.2-\frac{(0.2)^{2}}{2}+\frac{(0.2)^{3}}{3}=0.2-0.02+0.002\overline{6}\approx 0.18267. $$

(A1)

(c) Apply the Lagrange bound and check consistency(c) 代入拉格朗日误差界并验证一致性 M1·M1·A1·R1

Here $M=6$, $n=3$ (so $n+1=4$), $|x-a|=0.2$. (M1)此处 $M=6$,$n=3$(故 $n+1=4$),$|x-a|=0.2$。(M1)

$$ |R_3(1.2)|\le\frac{M}{4!}|x-a|^{4}=\frac{6}{24}(0.2)^{4}=0.25\times 0.0016=0.0004. $$

(M1) (A1)

The actual error is $|0.18267-0.18232|\approx 0.00035$, which is indeed $\le 0.0004$, confirming the bound is valid. (R1)实际误差为 $|0.18267-0.18232|\approx 0.00035$,确实 $\le 0.0004$,验证该误差界成立。(R1)

Insight.要点。 $M$ was found by noting $f^{(4)}(x)=-6/x^{4}$ is decreasing on $[1,1.2]$, so its maximum magnitude occurs at the left endpoint $x=1$, giving $M=6$; always check monotonicity of $|f^{(n+1)}|$ on the interval before just plugging in an endpoint. The Lagrange bound is deliberately conservative, it will always sit at or above the true error, never below, which is exactly what makes it useful as a guarantee.$M$ 的求法是注意到 $f^{(4)}(x)=-6/x^{4}$ 在 $[1,1.2]$ 上递减,故其绝对值的最大值出现在左端点 $x=1$,得 $M=6$;在直接代入某个端点之前,务必先检验 $|f^{(n+1)}|$ 在区间上的单调性。拉格朗日误差界故意取得保守:它总是大于等于真实误差,绝不会低估,这正是它能作为保证使用的原因。
FRQ 4HARD 10.13 Radius & Interval of Convergence10.13 收敛半径与收敛区间No Calculator[10 marks]

Let $\displaystyle\sum_{n=1}^{\infty}\frac{(x-2)^{n}}{n\cdot 5^{n}}$.设 $\displaystyle\sum_{n=1}^{\infty}\frac{(x-2)^{n}}{n\cdot 5^{n}}$。

Answers:答案:  (a) $R=5$  ·  (b) $x=-3$ converges; $x=7$ diverges$x=-3$ 收敛;$x=7$ 发散  ·  (c) $[-3,7)$  ·  (d) conditional条件收敛

(a) Ratio test on the power series(a) 对幂级数用比值判别法 M1·M1·A1

$\left|\dfrac{a_{n+1}}{a_n}\right|=\left|\dfrac{(x-2)^{n+1}}{(n+1)5^{n+1}}\cdot\dfrac{n\cdot 5^{n}}{(x-2)^{n}}\right|=\dfrac{|x-2|}{5}\cdot\dfrac{n}{n+1}$. (M1)$\left|\dfrac{a_{n+1}}{a_n}\right|=\left|\dfrac{(x-2)^{n+1}}{(n+1)5^{n+1}}\cdot\dfrac{n\cdot 5^{n}}{(x-2)^{n}}\right|=\dfrac{|x-2|}{5}\cdot\dfrac{n}{n+1}$。(M1)

Taking $n\to\infty$: $L=\dfrac{|x-2|}{5}$; convergence requires $L<1$. (M1)取 $n\to\infty$:$L=\dfrac{|x-2|}{5}$;收敛要求 $L<1$。(M1)

$|x-2|<5$, so $R=5$. (A1)$|x-2|<5$,故 $R=5$。(A1)

(b) Test each endpoint of $(-3,7)$(b) 检验 $(-3,7)$ 的每个端点 M1·A1·M1·A1

At $x=-3$: $(x-2)=-5$, so the term is $\dfrac{(-5)^{n}}{n\cdot 5^{n}}=\dfrac{(-1)^{n}}{n}$, the (negative of the) alternating harmonic series, which converges by the AST. (M1)(A1)在 $x=-3$ 处:$(x-2)=-5$,通项为 $\dfrac{(-5)^{n}}{n\cdot 5^{n}}=\dfrac{(-1)^{n}}{n}$,即交错调和级数(取相反数),由交错级数判别法收敛。(M1)(A1)

At $x=7$: $(x-2)=5$, so the term is $\dfrac{5^{n}}{n\cdot 5^{n}}=\dfrac{1}{n}$, the harmonic series, which diverges. (M1)(A1)在 $x=7$ 处:$(x-2)=5$,通项为 $\dfrac{5^{n}}{n\cdot 5^{n}}=\dfrac{1}{n}$,即调和级数,发散。(M1)(A1)

(c) Combine(c) 综合结论 A1

Include $x=-3$ (converges), exclude $x=7$ (diverges): the interval of convergence is $[-3,7)$. (A1)包含 $x=-3$(收敛),排除 $x=7$(发散):收敛区间为 $[-3,7)$。(A1)

(d) Classify the left endpoint(d) 判别左端点的收敛类型 M1·A1

At $x=-3$, the series of absolute values is $\displaystyle\sum\left|\frac{(-1)^{n}}{n}\right|=\sum\frac{1}{n}$, the harmonic series, which diverges. (M1)在 $x=-3$ 处,绝对值级数为 $\displaystyle\sum\left|\frac{(-1)^{n}}{n}\right|=\sum\frac{1}{n}$,即调和级数,发散。(M1)

Since the series converges at $x=-3$ (part (b)) but the absolute series diverges, the convergence there is conditional. (A1)因该级数在 $x=-3$ 处收敛((b) 部分)而绝对值级数发散,故该处为条件收敛。(A1)

Insight.要点。 The ratio test only ever produces the open interval, the two endpoints are always genuine toss-ups that require plugging the specific numeric value back into the original series and applying a completely different test (here, AST at one end, $p$-series at the other). A radius-of-convergence answer is incomplete without this endpoint work.比值判别法只能给出开区间,两个端点永远需要将具体数值代回原级数、再用完全不同的判别法(此处一端用交错级数判别法,另一端用 $p$ 级数判别法)单独检验,结果真正未知。没有端点检验的收敛半径答案是不完整的。
FRQ 5HARD 10.14 / 10.15 Maclaurin Series & Integration10.14 / 10.15 麦克劳林级数与积分Calculator[10 marks]

Let $g(x)=\ln(1+x)$.设 $g(x)=\ln(1+x)$。

Answers:答案:  (a) $\displaystyle\sum_{n=0}^{\infty}\frac{(-1)^{n}x^{n+1}}{n+1}$  ·  (b) $\ln(1.1)\approx 0.09531$  ·  (c) error误差 $\le 2\times 10^{-6}$  ·  (d) $R=1$; $x=1$ included属于收敛区间

(a) Substitute, then integrate term by term(a) 先代换,再逐项积分 M1·M1·A1

From $\dfrac{1}{1-u}=\sum u^{n}$, let $u=-x$: $\dfrac{1}{1+x}=\displaystyle\sum_{n=0}^{\infty}(-1)^{n}x^{n}$. (M1)由 $\dfrac{1}{1-u}=\sum u^{n}$,令 $u=-x$:$\dfrac{1}{1+x}=\displaystyle\sum_{n=0}^{\infty}(-1)^{n}x^{n}$。(M1)

Integrate term by term: $\displaystyle\int\frac{dx}{1+x}=\sum_{n=0}^{\infty}\frac{(-1)^{n}x^{n+1}}{n+1}+C$. (M1)逐项积分:$\displaystyle\int\frac{dx}{1+x}=\sum_{n=0}^{\infty}\frac{(-1)^{n}x^{n+1}}{n+1}+C$。(M1)

Since $g(0)=\ln 1=0$, evaluating the series at $x=0$ gives $C=0$, so $\ln(1+x)=\displaystyle\sum_{n=0}^{\infty}\frac{(-1)^{n}x^{n+1}}{n+1}=x-\frac{x^{2}}{2}+\frac{x^{3}}{3}-\cdots$. (A1)因 $g(0)=\ln 1=0$,将级数在 $x=0$ 处求值得 $C=0$,故 $\ln(1+x)=\displaystyle\sum_{n=0}^{\infty}\frac{(-1)^{n}x^{n+1}}{n+1}=x-\frac{x^{2}}{2}+\frac{x^{3}}{3}-\cdots$。(A1)

(b) Substitute $x=0.1$ and sum four terms(b) 代入 $x=0.1$ 并求四项之和 M1·A1

(M1)(M1)

$$ 0.1-\frac{(0.1)^{2}}{2}+\frac{(0.1)^{3}}{3}-\frac{(0.1)^{4}}{4}=0.1-0.005+0.0003\overline{3}-0.000025\approx 0.09531. $$

(A1)

(c) Bound the error by the fifth term(c) 用第五项界定误差 M1·A1

The first omitted term is the $n=4$ term: $\dfrac{x^{5}}{5}=\dfrac{(0.1)^{5}}{5}$. (M1)第一个被舍弃的项为 $n=4$ 项:$\dfrac{x^{5}}{5}=\dfrac{(0.1)^{5}}{5}$。(M1)

$\dfrac{0.00001}{5}=0.000002=2\times10^{-6}$, so the error in part (b) is at most $2\times10^{-6}$. (A1)$\dfrac{0.00001}{5}=0.000002=2\times10^{-6}$,故 (b) 中的误差至多为 $2\times10^{-6}$。(A1)

(d) Radius, then test the right endpoint(d) 求半径,再检验右端点 M1·M1·A1

The geometric series for $1/(1+x)$ converges for $|x|<1$, and term-by-term integration preserves the radius, so $R=1$. (M1)$1/(1+x)$ 的等比级数在 $|x|<1$ 时收敛,逐项积分保持半径不变,故 $R=1$。(M1)

At $x=1$: $\displaystyle\sum_{n=0}^{\infty}\frac{(-1)^{n}}{n+1}=1-\frac12+\frac13-\cdots$; $b_n=1/(n+1)$ is positive, decreasing, and $\to 0$, so this converges by the AST. (M1)在 $x=1$ 处:$\displaystyle\sum_{n=0}^{\infty}\frac{(-1)^{n}}{n+1}=1-\frac12+\frac13-\cdots$;$b_n=1/(n+1)$ 为正、递减且 $\to 0$,由交错级数判别法收敛。(M1)

So $x=1$ is included, and the interval of convergence for the series in (a) is $(-1,1]$. (A1)故 $x=1$ 属于收敛区间,(a) 中级数的收敛区间为 $(-1,1]$。(A1)

Insight.要点。 Term-by-term integration is a free way to generate a new series, but it is not a free way to extend where it converges: the radius survives, yet an endpoint that failed for $\sum x^{n}$ can succeed after integration, because integration divides each term by $n+1$, which is exactly the extra shrinkage that flips the harmonic series ($x=-1$ for $1/(1+x)$'s own series) into a convergent alternating harmonic series here at $x=1$.逐项积分是生成新级数的免费方法,但它不能免费地扩展收敛范围:半径不变,但对 $\sum x^{n}$ 失败的端点,积分后可能成功,因为积分将每项除以 $n+1$,这正是使调和级数($1/(1+x)$ 自身级数在 $x=-1$ 处)在此处 $x=1$ 转化为收敛的交错调和级数所需的额外收缩。