On 29 Aug 2003 at 19:31, Donald Burrill wrote:

Yes, you must give more details of what data you have. Depending, it 
might be possible to use some sort of EM-algorithm.

Kjetil Halvorsen

> Hi, Julianne.
> 
> What kind of "frequency distribution" do you have?  And what kind of
> linear model do you want to try to fit?  If your data consist wholly of
> univariate frequencies, it won't be possible to use it in a linear model
> (since a linear model implies at least two variables).  Similarly, if
> the model you want to pursue contains three (or more) variables, and you
> have only two-dimensional cross-tabulations, you cannot pursue that
> model.
> 
> On 29 Aug 2003, Julianne wrote (edited):
> 
> > Hi all,
> >
> > I have a statistical problem that I would greatly appreciate your
> > help with.
> >  I don't have individual observations, I have a frequency
> > distribution. I need to know how to get this data recognized by a
> > statistical program (statistica) to be used in a general linear
> > model.
> 
>  -----------------------------------------------------------------------
>  Donald F. Burrill                                         [EMAIL PROTECTED]
>  56 Sebbins Pond Drive, Bedford, NH 03110                 (603) 626-0816
> 
> .
> .
> =================================================================
> Instructions for joining and leaving this list, remarks about the
> problem of INAPPROPRIATE MESSAGES, and archives are available at:
> .                  http://jse.stat.ncsu.edu/                    .
> =================================================================


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