In article <[EMAIL PROTECTED]>,
David Delgado Gomez  <[EMAIL PROTECTED]> wrote:
>Good morning,

>I have data with a normal distribution.  Values higher than the mean are
>corrupted with noise. Is it possible to estimate the variance of the
>gaussian distribution just taking into account values smaller than the
>mean?
>Thanks
>David

It depends on how bad the noise is, and what is known about
the joint distribution of value and noise.  If the noise
does not make any above-mean values below the mean, just
using the sample median as an estimate of the mean and
estimating the variance by the mean square of the deviation
of the below-median observations from the median will
achieve half the efficiency of the no-noise situation.  As
usual, the specific assumptions matter.


-- 
This address is for information only.  I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
[EMAIL PROTECTED]         Phone: (765)494-6054   FAX: (765)494-0558
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to