Re: Determining if samples are normal

[Doug Stewart]

"normality test"   would be good.


For what it worth you can alter this slightly and get:

LogNormality test
Chisquare test
discrete test
etc.


Doug



Søren Hauberg wrote:

> Thanks. Does this test have a name or should I refer to it as a 
> Stewart-test :-) ?
>
> /Søren
>
> Doug Stewart wrote:
>
> > Søren Hauberg wrote:
> >
> > > Hi,
> > > Does anybody know how I can test wether or not some samples are 
> > > normaly distributed? I tried graphical methods, such as looking at 
> > > histograms and qqplots, but I don't trust my own judgement enough to 
> > > use graphical methods.
> > >
> > > /Søren
> >
> >
> >
> > Here is how I did it:
> > You warp it by the normal curve so that 'normal' data will produce a 
> > streight line.
> > Then you check how streight your line is with the cor() function..
> > A cor=1 is perfect fit.
> > A cor =.5 is there is a 50% chance it is from a normal distribution. 
> > etc.
> > The number of data points is also important. etc.
> > So if you get cor>.9 you can be happy.
> >
> >
> > ## Copyright (C) 2005 doug



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------