Timeline for Concentration bound for a increasingly weighted sum of bernoulli random variables
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 10 at 16:21 | vote | accept | Betty | ||
| May 10 at 11:44 | answer | added | Iosif Pinelis | timeline score: 1 | |
| May 9 at 14:32 | history | edited | Betty | CC BY-SA 4.0 |
added 21 characters in body
|
| May 9 at 14:32 | comment | added | Betty | Thanks for the comments, I have updated the question. For $c=1+1/m$ with $m\geq n$, how should I derive the concentration bound? | |
| May 9 at 12:40 | comment | added | Iosif Pinelis | If $c\ge2$ (say) or, more generally, if $c\ge1+t_n$ with $nt_n\to\infty$, then the weighted sum of the $n$ random variables $c^{i-1}x_i$ will be close to the sum of a few $c^{i-1}x_i$'s with the largest values of $i$, and then there will no concentration. That is, you can get concentration only if $c=1+O(1/n)$. | |
| S May 9 at 2:23 | review | First questions | |||
| May 9 at 5:47 | |||||
| S May 9 at 2:23 | history | asked | Betty | CC BY-SA 4.0 |