| Re: testing normality (& log normality) [message #42242] |
Fri, 14 January 2005 03:42 |
Paolo Grigis
Messages: 171
Registered: December 2003
|
Senior Member |
|
|
A simple method (for normality, anyway) would be to get the
distribution of your data (using histogram) and compare it with
a gaussian curve with the same average and width as your data.
Of course, there is some arbitrariness in the choice of the
histogram bin width, so some people won't like this kind of
approach and would use methods not requiring binning of the
original data.
Cheers,
Paolo
Mick Pope wrote:
> Greetings IDL gurus.
>
> I need to test how close to normal (and log normal) some distributions are.
> Is there a routine to do this?
>
> I note that the IDL curve fitting gives chi squared, DOF, etc, but is there
> something specifically suited to testing for the above fit somewhere?
>
> Cheers
>
> Mick...
>
>
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
