Skip to main content
fixed capital
Link
YCor
  • 63.9k
  • 5
  • 187
  • 286

When does the taylorTaylor coefficient of $e^{\sin x}$ vanish?

Source Link
T. Amdeberhan
  • 43.2k
  • 5
  • 57
  • 217

When does the taylor coefficient of $e^{\sin x}$ vanish?

If $f(x)=\frac{a_1}{1!}x+\frac{a_2}{2!}x^2+\frac{a_3}{3!}x^3+\frac{a_4}{4!}x^4+\cdots$ is an exponential generating function for $\{a_k\}_{k\geq1}$ then $$e^{f(x)}=1+\frac{a_1}{1!}x+\frac{a_1^2+a_2}{2!}x^2+\frac{a_1^3+3a_1a_2+a_3}{3!}x^3+\cdots$$ is related to the number of set partitions. A particular case $f(x)=e^x-1$ results in the Bell numbers via $e^{e^x-1}$.

Consider now the choice of $f(x)=\sin x$. I would like to ask:

QUESTION. Is it true that $b_n=0$ iff $n=3$ when $e^{\sin x}=\sum_{n\geq0}\frac{b_n}{n!}x^n$?