(07/14/2022, 08:37 AM)Catullus Wrote:(07/13/2022, 07:00 PM)JmsNxn Wrote: No, Daniel's right, there is no such function--at least no such analytic function. I doubt there's even a smooth function.How many times differentiable do you think it could be?
Every analytic function must satisfy:
\[
\lim_{n\to\infty} \left(\frac{f^{(n)}(k)}{n!}\right)^{1/n} \le 1
\]
And it's safe to say:
\[
\lim_{n\to\infty} \left(\frac{^n a}{n!}\right)^{1/n} = \infty\\
\]
Do you have any more guesses about that?
Radius zero implies not defined outsite its expansion point.
So nothing to take the derivative from.
unless you have another nontaylor definition that does compute values ofcourse.
regards
tommy1729
ps : a taylor expansion expanded at another point will not work since your derivatives diverges faster than exponential.
to see this : if f(x) has radius zero and derivatives diverges faster than exponential then f( c x ) has the same problem.
Plz think about it before you reply.
regards
tommy1729

