Timeline for Asymptotics of a Bernoulli-number-like function
Current License: CC BY-SA 2.5
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 13, 2011 at 20:28 | comment | added | Noam D. Elkies | But the exponent of $1-2^{-r} = 1 - 2^{-u}/n$ should be $n-1$, not $n$, right? I checked this for small $n$; it doesn't change the asymptotic oscillation but is needed to account for the discrepancy between the limiting oscillation and the numerical values for finite $n$. | |
| Jan 10, 2010 at 5:18 | history | edited | Reid Barton | CC BY-SA 2.5 |
added 1610 characters in body
|
| Jan 10, 2010 at 0:34 | comment | added | Michael Lugo | My computations agree with this. | |
| Jan 10, 2010 at 0:28 | history | answered | Reid Barton | CC BY-SA 2.5 |