Lessons and exercises follow the school system and stay in its language.Les leçons et les exercices suivent le système scolaire et restent dans sa langue.Lektionen und Übungen folgen dem Schulsystem und bleiben in seiner Sprache.
Every lesson of the chapter, ready for paper. Exercise sets are lettered A, B, C… and answered at the end.
To save a PDF, choose “Save as PDF” in the print dialog.
Calculus BC (AP-aligned)
Taylor and Maclaurin series
NicknameDate
Scan to open onlineenglish.sedifier.com/en-us/us-ma-2017/calculus-bc/us.calcbc.taylor/
1. Taylor polynomials
Warm-up
Three quick questions from earlier lessons.
f(2)=3 and f′(2)=2. Write the local linearization L(x) of f at x=2.
f(x)=x has f′(x)=2x1. Use the local linearization at x=4 to estimate 6. Give an exact fraction.
Near x=4, the graph of f is concave up. Does the local linearization at x=4 over- or underestimate f(x) for x near 4?
Underestimate: the tangent line lies below the curve.
Overestimate: the tangent line lies above the curve.
Neither: near x=4 the linearization is exact.
Discover
The violet curve is y=ex. The blue curve is 1+x+2x2+a6x3. Switch a from 0 to 1 to add the cubic term.
The curve y = eˣ and its Maclaurin polynomial of degree 2 (a = 0) or 3 (a = 1). The cubic term keeps the polynomial close to eˣ over a wider range.
What happens when you add the cubic term?
The polynomial hugs ex over a wider range
Nothing changes near the curve
Key idea
Match the value and derivatives at a point.
The Taylor polynomial of degree n of f centered at a is Pn(x)=k=0∑nk!f(k)(a)(x−a)k. Centered at 0 it is a Maclaurin polynomial.
Worked example
P3 of sinx at 0.
f(0)=0, f′(0)=1, f′′(0)=0, f(3)(0)=−1
Derivatives at 0: sin, cos, −sin, −cos.
P3(x)=x−6x3
Divide by k!.
P3(0.5)=0.4792
sin(0.5) = 0.4794…
Your turn, guided
Your turn: P2 of cosx at 0.
P2(x)=1−2x2
1, 0, −1 divided by 0!, 1!, 2!.
Practice ★
f(0)=−6, f′(0)=−6 and f′′(0)=4. Write the degree-2 Maclaurin polynomial of f.
f(0)=4, f′(0)=4 and f′′(0)=1. Write the degree-2 Maclaurin polynomial of f.
Find the degree-2 Maclaurin polynomial of f(x)=e8x.
Find the degree-2 Maclaurin polynomial of f(x)=e6x.
Watch out!
Practice ★★
Find the degree-2 Taylor polynomial of f(x)=8lnx centered at x=1.
Find the degree-2 Taylor polynomial of f(x)=6lnx centered at x=1.
Exit ticket
Three questions to check you have got it.
Find the degree-2 Maclaurin polynomial of f(x)=e8x.
Find the degree-2 Maclaurin polynomial of f(x)=e6x.
Find the degree-2 Maclaurin polynomial of f(x)=e7x.
Summary card
Taylor polynomials
Pn(x)=k=0∑nk!f(k)(a)(x−a)k.
Watch out: divide by k!; use powers of (x−a).
One more step is optional: a challenge. Try it, or finish the lesson now.
Challenge ★★★
Find the degree-2 Taylor polynomial of f(x)=x centered at x=1.
2. The Lagrange error bound
Warm-up
Three quick questions from earlier lessons.
Find the degree-2 Maclaurin polynomial of f(x)=e8x.
f(0)=−6, f′(0)=−6 and f′′(0)=4. Write the degree-2 Maclaurin polynomial of f.
Find the degree-2 Taylor polynomial of f(x)=8lnx centered at x=1.
Discover
P2 matches f, f′ and f′′ at a. Which derivative is responsible for the error?
f(3)
f′′
Key idea
The next term, with the worst derivative, bounds the error.
Lagrange error bound: if ∣f(n+1)(t)∣≤M between a and x, then ∣f(x)−Pn(x)∣≤(n+1)!M∣x−a∣n+1.
Worked example
Bound the error of P3 for ex at x=0.5 (center 0).
M=2
f⁽⁴⁾ = eˣ ≤ e^0.5 < 2 on [0, 0.5].
∣R3∣≤242⋅0.54
|x|⁴/4!.
≈0.0052
Compute.
Your turn, guided
Your turn: ∣f(3)∣≤6; bound the error of P2 at ∣x−a∣=1.
∣R2∣≤1
6 · 1³/3!.
Practice ★
∣f(3)(x)∣≤3 on [0,1]. Bound the error of the degree-2 Maclaurin polynomial at x=1.
∣f(3)(x)∣≤13 on [0,1]. Bound the error of the degree-2 Maclaurin polynomial at x=1.
The Lagrange error bound for the degree-4 Taylor polynomial uses the maximum of which derivative?
f(5), the next derivative
f(4)
f itself
Watch out!
Practice ★★
On the interval, ∣f(2)(x)∣≤2. Bound the error of the degree-1 Taylor polynomial centered at a, at a point with ∣x−a∣=21.
On the interval, ∣f(3)(x)∣≤1. Bound the error of the degree-2 Taylor polynomial centered at a, at a point with ∣x−a∣=21.
Exit ticket
Three questions to check you have got it.
∣f(3)(x)∣≤3 on [0,1]. Bound the error of the degree-2 Maclaurin polynomial at x=1.
∣f(3)(x)∣≤13 on [0,1]. Bound the error of the degree-2 Maclaurin polynomial at x=1.
∣f(3)(x)∣≤1 on [0,1]. Bound the error of the degree-2 Maclaurin polynomial at x=1.
Summary card
Lagrange error bound
∣Rn∣≤(n+1)!M∣x−a∣n+1, M≥∣f(n+1)∣.
Watch out: n+1 everywhere.
One more step is optional: a challenge. Try it, or finish the lesson now.
Challenge ★★★
Bound the error of the degree-3 Maclaurin polynomial of sinx at x=91, using ∣f(4)(x)∣≤1.
3. Radius and interval of convergence of a power series
Warm-up
Three quick questions from earlier lessons.
S=n=1∑∞n2(−1)n+12.
The partial sum S3: ?
A bound on ∣S−S3∣: ?
n=1∑∞n2(−1)n+12 is a p-series with p=?
nth-term check: n→∞limn22=?
The ratio test on ∑n22: L=? (so the test is inconclusive)
For an=2nn, find L=n→∞limanan+1.
For n=1∑∞n21, the ratio test gives L=1. What does it tell you?
Nothing: the ratio test is inconclusive when L = 1
The series converges
The series diverges
Discover
n=0∑∞xn is geometric. For which x does it converge?
−1<x<1
−1≤x≤1
Key idea
A power series converges in an interval around its center.
∑cn(x−a)n converges for ∣x−a∣<R and diverges for ∣x−a∣>R (the radius). Find R with the ratio test. At the endpoints x=a±R anything can happen: test each one separately.
Worked example
n=1∑∞n3nxn.
R=3
Ratio test: |x|/3 · n/(n + 1) → |x|/3.
Right endpoint excluded
x = 3: ∑1/n diverges.
Interval [−3,3)
x = −3: ∑(−1)ⁿ/n converges.
Your turn, guided
Your turn: the radius of ∑5nxn.
R=5
|x|/5 < 1.
Practice ★
Find the radius of convergence of n=0∑∞2nxn.
Find the radius of convergence of n=0∑∞27nxn.
Find the radius of convergence of n=0∑∞(2x)n.
Find the radius of convergence of n=0∑∞(66x)n.
Watch out!
Practice ★★
What is the interval of convergence of n=1∑∞n(x−7)n?
[6,8)
(6,8)
[6,8]
Exit ticket
Three questions to check you have got it.
Find the radius of convergence of n=0∑∞2nxn.
Find the radius of convergence of n=0∑∞27nxn.
Find the radius of convergence of n=0∑∞3nxn.
Summary card
Radius and interval
Ratio test → ∣x−a∣<R.
Test x=a−R and x=a+R separately.
Watch out: solve for ∣x−a∣ carefully; never skip the endpoints.
One more step is optional: a challenge. Try it, or finish the lesson now.
Challenge ★★★
Find the radius of convergence of n=1∑∞n⋅81nxn.
4. Maclaurin series of eˣ, sin x, cos x and 1/(1 − x)
Warm-up
Three quick questions from earlier lessons.
Find the degree-2 Maclaurin polynomial of f(x)=e8x.
f(0)=−6, f′(0)=−6 and f′′(0)=4. Write the degree-2 Maclaurin polynomial of f.
Find the degree-2 Taylor polynomial of f(x)=8lnx centered at x=1.
Discover
Every derivative of ex at 0 is 1. So its Maclaurin series is ∑n!xn. What about sinx, whose derivatives at 0 cycle 0,1,0,−1?
Find the coefficient of x3 in the Maclaurin series of e5x.
Find the coefficient of x3 in the Maclaurin series of e6x.
n=0∑∞(2n+1)!(−1)n(1x)2n+1 is the Maclaurin series of which function?
sin(1x)
cos(1x)
e1x
Watch out!
Practice ★★
Find the coefficient of x4 in the Maclaurin series of cos(2x).
Find the coefficient of x4 in the Maclaurin series of cos(5x).
Exit ticket
Three questions to check you have got it.
Find the coefficient of x3 in the Maclaurin series of e5x.
Find the coefficient of x3 in the Maclaurin series of e6x.
Find the coefficient of x3 in the Maclaurin series of e4x.
Summary card
Known Maclaurin series
ex, sinx (odd), cosx (even), 1−x1.
Watch out: signs alternate in sin and cos.
One more step is optional: a challenge. Try it, or finish the lesson now.
Challenge ★★★
Find the coefficient of x5 in the Maclaurin series of sin(4x).
5. New series from known ones
Warm-up
Three quick questions from earlier lessons.
Find the antiderivative F of f(x)=3x2+2 with F(0)=0.
Find the antiderivative F of f(x)=cosx+ex with F(0)=5.
Find the antiderivative F of f(x)=x1, for x>0, with F(1)=−9.
Discover
To get the series of xsinx, can you just multiply the series of sinx by x?
Yes, each power goes up by one
No, the derivatives must be computed again
Key idea
Build new series from known ones.
Inside its interval, a power series can be multiplied byxk, have xreplaced by u(x) (such as −x2 or 3x), and be differentiated or integrated term by term. The radius stays the same for differentiation and integration.
Worked example
The series of ln(1+x).
1+x1=1−x+x2−x3+⋯
Substitute −x into 1/(1 − u).
ln(1+x)=x−2x2+3x3−⋯
Integrate term by term from 0.
Your turn, guided
Your turn: the coefficient of x3 in x2ex.
1
x² · x¹/1!.
Practice ★
Find the coefficient of x3 in the Maclaurin series of xe4x.
Find the coefficient of x3 in the Maclaurin series of xe9x.
Find the coefficient of x3 in the Maclaurin series of xe2x.
Watch out!
Practice ★★
Find the coefficient of x10 in the Maclaurin series of 1+x21.
Find the coefficient of x2 in the Maclaurin series of 1+x21.
arctanx=∫0x1+t2dt. Find the coefficient of x3 in its Maclaurin series.
arctanx=∫0x1+t2dt. Find the coefficient of x9 in its Maclaurin series.
Exit ticket
Three questions to check you have got it.
Find the coefficient of x3 in the Maclaurin series of xe4x.
Find the coefficient of x3 in the Maclaurin series of xe9x.
Find the coefficient of x3 in the Maclaurin series of xe2x.
Summary card
New series from old
Multiply by xk: shift the powers.
Substitute: replace x by u(x).
Differentiate or integrate term by term.
Watch out: integrating divides by the new power.
One more step is optional: a challenge. Try it, or finish the lesson now.
Challenge ★★★
Integrate the series of 1+x1 to get ln(1+x). Find the coefficient of x6.
6. Fibonacci sequences and the golden ratio
Discover
1, 1, 2, 3, 5, 8, 13, … each term is the sum of the two before. Divide each term by the one before: 2, 1.5, 1.67, 1.6, 1.625, … Do these ratios settle down?
Yes, near 1.618
No, they keep growing like the terms
Key idea
Fibonacci and the golden ratio.
If un+1=un+un−1 and the ratio unun+1 tends to x, then x=1+x1, so x2=x+1 and
x=φ=21+5≈1.618.
Worked example
A sequence follows the same rule with u1=2 and u2=5. Find u5.
u3=7, u4=12
u3=5+2, u4=7+5.
u5=19
u5=12+7.
Your turn, guided
Your turn: with u1=1 and u2=3, what is u4?
u4=7
u3=4, then u4=4+3.
Practice ★
un+1=un+un−1 with u1=3 and u2=6. Find u5.
un+1=un+un−1 with u1=4 and u2=4. Find u5.
un+1=un+un−1 with u1=5 and u2=6. Find u5.
Practice ★★
The ratio of consecutive Fibonacci numbers tends to the positive solution of x2=x+1. Give it exactly.
Exit ticket
Two questions to check.
un+1=un+un−1 with u1=3 and u2=6. Find u5.
un+1=un+un−1 with u1=4 and u2=4. Find u5.
Summary card
Fibonacci-type sequences
Each term is the sum of the two before.
The ratio of consecutive terms tends to φ≈1.618.
Spirals in sunflowers and pine cones often show Fibonacci numbers.
7. History of sequences: Zeno, Fibonacci and Gauss
Discover
Around 450 BC, Zeno of Elea argued: to cross a room you must first reach the middle, then the middle of what is left, and so on, forever. So you can never arrive! What is wrong?
Infinitely many steps can have a finite total: 21+41+81+⋯=1
Nothing: motion is impossible
Key idea
Sequences through history.
Step 1Zeno of Eleac. 490 – c. 430 BCImagined portraitJan de Bisschop, 1666–1671. CC0 1.0Zeno of Elea (about 490–425 BC): the dichotomy, an infinite number of halves.
Step 2Fibonaccic. 1170 – c. 1250Imagined portraitUnknown artist. Public domain1202: Fibonacci's Liber abaci brings Hindu-Arabic numerals to Europe and poses the rabbit problem.
Step 3Carl Friedrich Gauss1777–1855Gottlieb Biermann / After Christian Albrecht Jensen, 1840. Public domainGauss (born 1777), at school, adds 1 to 100 instantly by pairing the numbers.
Zeno's puzzles led to infinite sums. In 1202 Fibonacci's Liber abaci asked how many pairs of rabbits grow from one pair: 1,1,2,3,5,8,… As a schoolboy, Gauss added 1+2+⋯+100 at once: 50 pairs, each summing to 101.
Worked example
Gauss's trick for 1+2+⋯+100:
each pair is 101
Pair the first and last terms: 1+100, 2+99, …
50×101=5050
There are 50 pairs.
Practice ★
Besides the rabbit problem, what did Fibonacci's Liber abaci (1202) bring to Europe?
The Hindu-Arabic place-value numerals
Logarithm tables
Roman numerals
Use Gauss's pairing trick: 1+2+⋯+98.
Use Gauss's pairing trick: 1+2+⋯+78.
Zeno's walker has covered 21+41+⋯+231 of the room. How much is that?
Summary card
Sequences today
Savings and loans grow as geometric sequences.
Computer programs repeat steps with recursive rules, like Fibonacci's.
Zeno's sums appear whenever a quantity halves again and again, such as radioactive decay.
Sources: MacTutor biographies of Zeno of Elea, Fibonacci and Gauss (see "Sources and credits").
Answers: Taylor and Maclaurin series
1. Taylor polynomials
A
2x−1
25
Underestimate: the tangent line lies below the curve.
B
2x2−6x−6
2x2+4x+4
C
32x2+8x+1
18x2+6x+1
D
8x−4(x−1)2−8
6x−3(x−1)2−6
E
32x2+8x+1
18x2+6x+1
249x2+7x+1
F
2x−8(x−1)2+21
2. The Lagrange error bound
A
32x2+8x+1
2x2−6x−6
8x−4(x−1)2−8
B
21
613
C
f(5), the next derivative
D
41
481
E
21
613
61
F
1574641
3. Radius and interval of convergence of a power series
A
a) 1831 b) 81 c) 2 d) 0 e) 1
21
Nothing: the ratio test is inconclusive when L = 1
B
2
27
C
21
661
D
[6,8)
E
2
27
3
F
81
4. Maclaurin series of eˣ, sin x, cos x and 1/(1 − x)
A
32x2+8x+1
2x2−6x−6
8x−4(x−1)2−8
B
6125
36
C
sin(1x)
D
32
24625
E
6125
36
332
F
15128
5. New series from known ones
A
x3+2x
ex+sin(x)+4
ln(x)−9
B
8
281
2
C
−1
−1
D
−31
91
E
8
281
2
F
−61
6. Fibonacci sequences and the golden ratio
A
24
20
28
B
21+25
C
24
20
7. History of sequences: Zeno, Fibonacci and Gauss