Infinite Differentiability and Analyticity.

In summary, the function $f(x)=e^{-1/x^2}$ is infinitely differentiable but not analytic at the point $x=0$. This is because while all of its derivatives are continuous and equal to zero at $x=0$, the function itself is not zero for $x \neq 0$, leading to a contradiction that it can be written as a pointwise convergent Taylor series on any open interval containing $x=0$. This implies that for any $M, \delta > 0$, in a $\delta$-neighbourhood of the origin, the function $f$ satisfies $|f^{(k)}(x)| > M\cdot k!/\delta^k$.
  • #1
Alone
60
0
I want to show that the function defined as follows:

$f(x)=e^{-1/x^2}$ for $|x|>0$ and $f^{(k)}(0)=0$ for $k=0,1,2,\ldots$ is infinitely differentiable but not analytic at the point $x=0$.

For infinite-differentiability I used the fact that $\lim_{|x|\to 0^+} x^{-n} e^{-1/x^2}=0$ for every $n$ nonnegative integer, from which I can deduce that this function is infinite differnetiable at $x=0$ since all of its derivatives are continuous at $x=0$ and equal zero.

The problem I am having is with showing that it's not analytic, I need to show that for every $\delta>0, M>0$ the derivative $f^{(k)}(x)$ its absolute value is greater than $M\cdot k!/\delta^k$; not sure how to show it exactly.
 
Physics news on Phys.org
  • #2
To start by picking a nit, I would like to remark that the equality $f^{(k)}(0) = 0$ for $k \ge 1$ is not part of the definition of $f$, but a consequence of the definition of $f$.

If $f$ were analytic, there would exist $r > 0$ such that $f$ can be written as a pointwise convergent Taylor series on the open interval $(-r,r)$. Now, $f(0) = 0$ and you wrote that you have proven that all derivatives of $f$ at the origin vanish. However, $f(x) \neq 0$ for any $x \neq 0$. So what could you conclude?
 
  • #3
Krylov said:
To start by picking a nit, I would like to remark that the equality $f^{(k)}(0) = 0$ for $k \ge 1$ is not part of the definition of $f$, but a consequence of the definition of $f$.

If $f$ were analytic, there would exist $r > 0$ such that $f$ can be written as a pointwise convergent Taylor series on the open interval $(-r,r)$. Now, $f(0) = 0$ and you wrote that you have proven that all derivatives of $f$ at the origin vanish. However, $f(x) \neq 0$ for any $x \neq 0$. So what could you conclude?

I would get a contradiction since the Taylor series is identically zero, but the function isn't zero for $x\ne 0 $.

Thanks.
 
  • #4
@Krylov I have a follow up question:

I want to verify that for every $M,\delta>0$ in any $\delta$-neighbourhood of the origin the function $f$ that was given in my first post in this thread satisfies:

$$|f^{(k)}(x)| >Mk!/\delta^k$$

This can be concluded from the fact that $f$ isn't analytic, but can you show this explicitly?
 

1. What is the difference between infinite differentiability and analyticity?

Infinite differentiability refers to the property of a function that it can be differentiated an infinite number of times. Analyticity, on the other hand, refers to the property of a function that it can be represented by a power series expansion in a specific region. A function that is infinitely differentiable may or may not be analytic.

2. How can you determine if a function is infinitely differentiable or analytic?

A function can be determined to be infinitely differentiable by checking that all its derivatives exist and are continuous. To determine if a function is analytic, we can use the Cauchy-Riemann equations or the Cauchy Integral Theorem to check for complex differentiability.

3. Can a function be infinitely differentiable but not analytic?

Yes, a function can be infinitely differentiable but not analytic. For example, the function f(x) = |x| is infinitely differentiable but not analytic at the point x = 0.

4. What are some real-life applications of infinite differentiability and analyticity?

Infinite differentiability and analyticity are important concepts in calculus and complex analysis, which have numerous applications in physics, engineering, and economics. For example, in physics, these concepts are used to model physical phenomena such as motion and fluid flow. In economics, they are used in optimization problems and financial modeling.

5. Are there any limitations to infinite differentiability and analyticity?

Yes, there are some limitations to infinite differentiability and analyticity. For example, not all functions are infinitely differentiable or analytic. Also, for functions that are infinitely differentiable or analytic, the domains of their differentiability and analyticity may be limited. Additionally, these concepts may not always be applicable to non-smooth or discontinuous functions.

Similar threads

Replies
1
Views
951
Replies
2
Views
799
Replies
2
Views
897
Replies
11
Views
2K
Replies
14
Views
1K
Replies
5
Views
404
Replies
1
Views
1K
Replies
10
Views
3K
Back
Top