| |
 |
|
|
Science Forum Index » Statistics - Education Forum » Estimating variance of subsample given only variance...
Page 1 of 1
|
| Author |
Message |
| ... |
 Posted: Thu May 22, 2008 3:17 am |
|
|
|
Guest
|
Is there any estimate for a variance of the subset of N-samples? Say I
have 1000 samples and I only know the mean and variance, what would
the variance of the 100 sub samples be? |
|
|
| Back to top |
|
| Ray Koopman... |
| Posted: Thu May 22, 2008 12:00 pm |
|
|
|
Guest
|
On May 22, 6:17 am, mrmaxx2 at (no spam) gmail.com wrote:
Quote: Is there any estimate for a variance of the subset of N-samples? Say I
have 1000 samples and I only know the mean and variance, what would
the variance of the 100 sub samples be?
Regarding the variance of a subset of n values drawn randomly without
replacement from a finite population of N values, the best you can say
is that the expected value of the subset variance (i.e., the mean of
all possible N_choose_n subset variances) equals the population
variance, provided you calculate the population and subsets variances
using denominators of N-1 and n-1, respectively. |
|
|
| Back to top |
|
| |
|
Page 1 of 1
All times are GMT - 5 Hours
The time now is Sun Oct 12, 2008 11:18 am
|
|