Hi

On Tue, 1 Apr 2003 [EMAIL PROTECTED] wrote:

> Is it possible to generate a sample that has a specific mean, SD (also
> skewness, kurtosis)?
> To be more specific, I do not want to sample from a given distribution
> (like the normal distribution) that has a specific mean and SD. If
> sampled from a normal distribution with a given mean and SD, of course
> the mean of the sample is not the exact same as the mean of the normal
> distribution.

Seems straightforward with respect to mean and sd, at least if
you are willing to compute mean and sd for the initial sample of
scores and scale the scores to 0,1. I I'm not sure about
skewness and kurtosis.  In SPSS, something like (with calculation
of MNzy and SDzy determined elsewhere.

SET WIDTH = 80  SEED = 9000000
INPUT PROGRAM
.  LOOP subj = 1 TO 200
.    END CASE
.    END LOOP
.  END FILE
.  END INPUT PROGRAM
COMPUTE zy = NORM(1)
COMPUTE zy = (zy-mnzy)/sdzy  /* mnzy, sdzy computed elsewhere
DESCR zy  /* mean and sd = 0,1
COMP y = 100 + 10*zy
DESCR y  /* mean and sd = 100,10

The computation of mnzy and sdzy could be automated using the
AGGREGATE command or by generating multiple observations per case
(rather than multiple cases) and using

COMPUTE mnzy = MEAN(z1 to zn)
COMPUTE sdzy = SD(z1 to zn)

Best wishes
Jim

============================================================================
James M. Clark                          (204) 786-9757
Department of Psychology                (204) 774-4134 Fax
University of Winnipeg                  4L05D
Winnipeg, Manitoba  R3B 2E9             [EMAIL PROTECTED]
CANADA                                  http://www.uwinnipeg.ca/~clark
============================================================================

.
.
=================================================================
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