Power series solution for Log(1+x)

  • MHB
  • Thread starter ssh
  • Start date
  • #1
ssh
17
0
Show that,

\[\log(1+x)=x-\frac{x^2}{2}+\frac{x^3}{3}+\cdots\]
 
Last edited by a moderator:

Answers and Replies

  • #2
Fernando Revilla
Gold Member
MHB
631
0
Show that log(1+x) = x - x2\2 + x3​\3................

Use that $\dfrac{1}{1+x}=1-x+x^2-x^3+\ldots\quad (|x|<1)$ and take into account that all power series can be integrated term by term on an interval lying inside the interval of convergence.
 
  • #3
ssh
17
0
Can we write this as a Taylor's series as f(x) = Log(1+x), then f'(x)=1\1+x so on.
 
  • #4
Fernando Revilla
Gold Member
MHB
631
0
Can we write this as a Taylor's series as f(x) = Log(1+x), then f'(x)=1\1+x so on.

Of course you can. But using that method you only obtain the Taylor series of $f(x)=\log (1+x)$, that is $\displaystyle\sum_{n=1}^{\infty}(-1)^{n-1}\dfrac{x^n}{n}$. To prove that $\log (1+x)=\displaystyle\sum_{n=1}^{\infty}(-1)^{n-1}\dfrac{x^n}{n}$ in $(-1,1)$ (also in $x=1$) you need to verify that the remainder of the Taylor series converges to $0$.
 
  • #5
Prove It
Gold Member
MHB
1,466
24
Show that,

\[\log(1+x)=x-\frac{x^2}{2}+\frac{x^3}{3}+\cdots\]

If you want to use the long method, remember that a Maclaurin series for a function is given by $\displaystyle \begin{align*} f(x) = \sum_{n = 0}^{\infty} \frac{f^{(n)}(0)}{n!}x^n \end{align*}$

So evaluating the derivatives gives

$\displaystyle \begin{align*} f(x) &= \ln{(1 + x)} \\ f(0) &= 0 \\ f'(x) &= \frac{1}{1 + x} \\ f'(0) &= 1 \\ f''(x) &= -\frac{1}{(1 + x)^2} \\ f''(0) &= -1 \\ f'''(x) &= \frac{2}{(1 +x)^3} \\ f'''(0) &= 2 \\ f^{(4)}(x) &= -\frac{3!}{(1 + x)^4} \\ f^{(4)}(0) &= -3! \\ f^{(5)}(x) &= \frac{4!}{(1 + x)^5} \\ f^{(5)}(0) &= 4! \\ \vdots \end{align*}$

So substituting these in gives

$\displaystyle \begin{align*} \ln{(1 + x)} &= \frac{0}{0!}x^0 + \frac{1}{1!}x^1 + \frac{(-1)}{2!}x^2 + \frac{2}{3!}x^3 - \frac{3!}{4!}x^4 + \frac{4!}{5!}x^5 - \dots + \dots \\ &= x - \frac{x^2}{2} + \frac{x^3}{3} - \frac{x^4}{4} + \frac{x^5}{5} - \dots + \dots \end{align*}$
 
  • #6
Fernando Revilla
Gold Member
MHB
631
0
Now, we have to prove the Taylor's remainder [tex]R_n(x)[/tex]:

$$R_n(x)=\dfrac{f^{(n+1)}(\xi)}{(n+1)!}x^n=\dfrac{(-1)^n}{(1+\xi)^{n+1}}\dfrac{x^{n+1}}{(n+1)!}\quad (\xi \mbox{ between }0\mbox{ and }x)$$

has limit [tex]0[/tex] for [tex]x\in (-1,1)[/tex] as [tex]n\to \infty[/tex]. For that reason is better to use the series expansion of [tex]1/(1+x)[/tex].
 
  • #7
Saknussemm
4
0
Taking the derivative of the MacLaurin series gives you
$1 -x +x^2 - x^3 + x^4 + \ldots$
Since this is a geometric series with ratio $-x$, it equals $\frac{1}{1 + x}$ when x is in $(-1, 1)$.
This shows the expression $\ln(1+x)$ and its MacLaurin expansion to have the same derivative over $(-1, 1)$, which means they are equal within a constant. And, since they are equal at $x=0$, this constant is zero.

If my reasoning is correct, this is simpler than proving the limit of the Taylor remainder.
 
  • #8
HallsofIvy
Science Advisor
Homework Helper
43,021
970
Taking the derivative of the MacLaurin series gives you
$1 -x +x^2 - x^3 + x^4 + \ldots$
Since this is a geometric series with ratio $-x$, it equals $\frac{1}{1 + x}$ when x is in $(-1, 1)$.
This shows the expression $\ln(1+x)$ and its MacLaurin expansion to have the same derivative over $(-1, 1)$, which means they are equal within a constant. And, since they are equal at $x=0$, this constant is zero.

If my reasoning is correct, this is simpler than proving the limit of the Taylor remainder.

It is, however, the essentially the same as Fernando Revilla's suggestion in the first response to this thread.
 
  • #9
Saknussemm
4
0
It is, however, the essentially the same as Fernando Revilla's suggestion in the first response to this thread.

The logic and context are not the same, as I was answering the question whether you can use the Taylor series. In general, you prove the validity of the Taylor expansion over a given interval by proving the Taylor reminder tends to zero as n goes to infinity. But here, because we can bring out a geometric series, as Fernando Revilla does in the initial answer, we have a simpler alternative. It is actually an exceptional case that is worth noting.
 

Suggested for: Power series solution for Log(1+x)

  • Last Post
Replies
3
Views
8K
Replies
2
Views
3K
  • Last Post
Replies
13
Views
781
  • Last Post
Replies
3
Views
7K
Replies
19
Views
339
Replies
3
Views
962
  • Last Post
Replies
8
Views
49K
  • Last Post
Replies
2
Views
10K
Replies
2
Views
3K
  • Last Post
Replies
7
Views
7K
Top