ileslie
02-22-2008, 11:05 PM
I made up this series for a class the other day. It is a sum for k = 1,2,3, ... inf
of the terms (x/k)^k. It gets big rather quickly but will always converge for x >=0.
Once k > x the terms get small very quickly. Here is my question. Does anyone recognize this series? I have not been able to find it discussed online or in texts in the library. I am curious if it represents some known function(s). If not, has some asymptotic analysis been done. The sum seems to approach sqrt(c*x)*exp(x/e) for large x (x>400), with c close to 2.3104048729. This was found with Matlab, but I've made no serious attempt at any analysis.
Thanks in advance.
of the terms (x/k)^k. It gets big rather quickly but will always converge for x >=0.
Once k > x the terms get small very quickly. Here is my question. Does anyone recognize this series? I have not been able to find it discussed online or in texts in the library. I am curious if it represents some known function(s). If not, has some asymptotic analysis been done. The sum seems to approach sqrt(c*x)*exp(x/e) for large x (x>400), with c close to 2.3104048729. This was found with Matlab, but I've made no serious attempt at any analysis.
Thanks in advance.