微积分I(标准路径) · 免费预览

§8.4 Taylor series

函数 $f$ 在 $a$ 处的泰勒级数是什么?

1. Introduction

Introduction

2. Taylor series and the Ratio Test

Taylor series and the Ratio Test

3. Taylor series of several important functions

Taylor series of several important functions

4. Summary

Summary

5. Reading $a$ and $r$ from a series

The series T(x)=3−x+13x2−19x3+⋯T(x) = 3-x+\frac{1}{3}x^2-\frac{1}{9}x^3+\cdots is geometric with

6. The limiting ratio for a logarithm series

For the series of f(x)=ln⁡(2)−ln⁡(3−x)f(x) = \ln(2)-\ln(3-x) centered at 11, consecutive terms have ratios rn(x)=nn+1⋅12(x−1)r_n(x) = \frac{n}{n+1}\cdot\frac{1}{2}(x-1). The limit r(x)r(x) equals

7. The interval behind the ratio

For the logarithm series above with limiting ratio r(x)=12(x−1)r(x) = \frac{1}{2}(x-1), solving the convergence condition gives

8. The exponential ratio collapses to zero

For the Taylor series of exe^x, the term ratio is rn(x)=xn+1r_n(x) = \frac{x}{n+1}, so the limiting ratio is r(x)=0r(x) = 0. The Ratio Test then guarantees

9. The sine ratio shrinks to zero

For the sine series x−x33!+x55!−⋯x-\frac{x^3}{3!}+\frac{x^5}{5!}-\cdots, consecutive term ratios satisfy rn(x)=−x2(2n+2)(2n+3)r_n(x) = -\frac{x^2}{(2n+2)(2n+3)}, whose limit is 00. The Ratio Test therefore yields

10. Guessing the function from its polynomial

Plots show the degree 1010 Taylor polynomial of a mystery function matching it to pixel accuracy on the interval (1,3)(1,3), centered at 11. A natural conjecture is

11. The exponential from its series

The series ∑k=0∞xkk!\sum_{k=0}^{\infty}\frac{x^k}{k!} is the Taylor series of

12. The ratio of the exponential series

For the series ∑k=0∞xkk!\sum_{k=0}^{\infty}\frac{x^k}{k!}, the ratio of consecutive terms gives

13. Substituting into the sine series

Substituting x2x^2 into the sine series gives sin⁡(x2)=x2−x63!+x105!−⋯\sin(x^2) = x^2-\frac{x^6}{3!}+\frac{x^{10}}{5!}-\cdots, which converges

14. An inverted series at a new center

For f(x)=11+xf(x) = \frac{1}{1+x} centered at a=1a = 1, writing f(x)=12⋅11+x−12f(x) = \frac{1}{2}\cdot\frac{1}{1+\frac{x-1}{2}} gives the series

15. A Lagrange bound for the exponential

Using the degree 1010 polynomial of exe^x at a=0a=0 to estimate e2e^2, the Lagrange bound with M=9M = 9 gives error at most 9⋅21111!\frac{9\cdot2^{11}}{11!}. Report this bound, rounded to six decimals.

16. Degrees needed for cosine precision

To estimate cos⁡(1)\cos(1) within 0.000000010.00000001 using the Taylor polynomial at 00, the Lagrange bound with M=1M = 1 requires degree n=11n = 11. Report the required degree.

17. A table of unit derivatives

A spreadsheet table listing derivatives at zero shows every value f(k)(0)=1f^{(k)}(0) = 1 for the function behind the series ∑xkk!\sum\frac{x^k}{k!}.

18. Polynomial graphs hug an everywhere-convergent series

For the series behind these tables, the degree 1010 polynomial is visually indistinguishable from the function across −4<x<4-4 \lt x \lt 4, while degree 2020 covers roughly −8<x<8-8 \lt x \lt 8.

19. The cosine expansion at its peak

For f(x)=cos⁡(x)f(x) = \cos(x) centered at a=π2a = \frac{\pi}{2}, the degree 44 Taylor polynomial contains no (x−π2)2(x-\frac{\pi}{2})^2 term.

进入交互式练习

免费注册 · 即时批改 · 自动安排间隔复习。

免费开始这门课