Answer:
The Taylor series is
.
The radius of convergence is
.
Step-by-step explanation:
<em>The Taylor expansion.</em>
Recall that as we want the Taylor series centered at
its expression is given in powers of
. With this in mind we need to do some transformations with the goal to obtain the asked Taylor series from the Taylor expansion of
.
Then,
![\ln(x) = \ln(x-3+3) = \ln(3(\frac{x-3}{3} + 1 )) = \ln 3 + \ln(1 + \frac{x-3}{3}).](https://tex.z-dn.net/?f=%5Cln%28x%29%20%3D%20%5Cln%28x-3%2B3%29%20%3D%20%5Cln%283%28%5Cfrac%7Bx-3%7D%7B3%7D%20%2B%201%20%29%29%20%3D%20%5Cln%203%20%2B%20%5Cln%281%20%2B%20%5Cfrac%7Bx-3%7D%7B3%7D%29.)
Now, in order to make a more compact notation write
. Thus, the above expression becomes
![\ln(x) = \ln 3 + \ln(1+y).](https://tex.z-dn.net/?f=%5Cln%28x%29%20%3D%20%5Cln%203%20%2B%20%5Cln%281%2By%29.)
Notice that, if x is very close from 3, then y is very close from 0. Then, we can use the Taylor expansion of the logarithm. Hence,
![\ln(x) = \ln 3 + \ln(1+y) = \ln 3 + \sum_{n=1}^{\infty} (-1)^{n+1} \frac{y^n}{n}.](https://tex.z-dn.net/?f=%5Cln%28x%29%20%3D%20%5Cln%203%20%2B%20%5Cln%281%2By%29%20%3D%20%5Cln%203%20%2B%20%5Csum_%7Bn%3D1%7D%5E%7B%5Cinfty%7D%20%28-1%29%5E%7Bn%2B1%7D%20%5Cfrac%7By%5En%7D%7Bn%7D.)
Now, substitute
in the previous equality. Thus,
![\ln(x) = \ln 3 + \sum_{n=1}^{\infty} (-1)^{n+1} \frac{(x-3)^n}{3^n n}.](https://tex.z-dn.net/?f=%5Cln%28x%29%20%3D%20%5Cln%203%20%2B%20%5Csum_%7Bn%3D1%7D%5E%7B%5Cinfty%7D%20%28-1%29%5E%7Bn%2B1%7D%20%5Cfrac%7B%28x-3%29%5En%7D%7B3%5En%20n%7D.)
<em>Radius of convergence.</em>
We find the radius of convergence with the Cauchy-Hadamard formula:
![R^{-1} = \lim_{n\rightarrow\infty} \sqrt[n]{|a_n|},](https://tex.z-dn.net/?f=R%5E%7B-1%7D%20%3D%20%5Clim_%7Bn%5Crightarrow%5Cinfty%7D%20%5Csqrt%5Bn%5D%7B%7Ca_n%7C%7D%2C)
Where
stands for the coefficients of the Taylor series and
for the radius of convergence.
In this case the coefficients of the Taylor series are
![a_n = \frac{(-1)^{n+1}}{ n3^n}](https://tex.z-dn.net/?f=%20a_n%20%3D%20%5Cfrac%7B%28-1%29%5E%7Bn%2B1%7D%7D%7B%20n3%5En%7D)
and in consequence
. Then,
![\sqrt[n]{|a_n|} = \sqrt[n]{\frac{1}{3^nn}}](https://tex.z-dn.net/?f=%20%5Csqrt%5Bn%5D%7B%7Ca_n%7C%7D%20%3D%20%5Csqrt%5Bn%5D%7B%5Cfrac%7B1%7D%7B3%5Enn%7D%7D)
Applying the properties of roots
![\sqrt[n]{|a_n|} = \frac{1}{3\sqrt[n]{n}}.](https://tex.z-dn.net/?f=%20%5Csqrt%5Bn%5D%7B%7Ca_n%7C%7D%20%3D%20%5Cfrac%7B1%7D%7B3%5Csqrt%5Bn%5D%7Bn%7D%7D.)
Hence,
![R^{-1} = \lim_{n\rightarrow\infty} \frac{1}{3\sqrt[n]{n}} =\frac{1}{3}](https://tex.z-dn.net/?f=R%5E%7B-1%7D%20%3D%20%5Clim_%7Bn%5Crightarrow%5Cinfty%7D%20%5Cfrac%7B1%7D%7B3%5Csqrt%5Bn%5D%7Bn%7D%7D%20%3D%5Cfrac%7B1%7D%7B3%7D)
Recall that
![\lim_{n\rightarrow\infty} \sqrt[n]{n}=1.](https://tex.z-dn.net/?f=%5Clim_%7Bn%5Crightarrow%5Cinfty%7D%20%5Csqrt%5Bn%5D%7Bn%7D%3D1.)
So, as
we get that
.