Page 1 of 1

Normtest

Posted: Wed Sep 08, 2010 12:20 pm
by EViews Esther
This thread is about the Normtest add-in that includes many of the dozens of techniques for assessing univariate/multivariate normality.

Re: Normtest

Posted: Tue Sep 13, 2011 8:55 pm
by lucem_fert
Hi!

I'Ve downloaded this Add-in but i don't understand how i can use it. Anyone knows it?

Thanks.

Normtest

Posted: Tue Sep 13, 2011 9:32 pm
by EViews Gareth
I'm sure it has documentation, available from the manage add-ins menu.

Re: Normtest

Posted: Thu Sep 15, 2011 12:41 am
by lucem_fert
Yes you're right!!

It was very very late in the night when i was studyng it!

Re: Normtest

Posted: Mon Jan 09, 2012 1:28 pm
by JamieLevon
I can't access the documentation. Where is the add-ins menu?

Re: Normtest

Posted: Mon Jan 09, 2012 2:23 pm
by EViews Gareth
Between the Options menu and the Window Menu in EViews 7.2.

Re: Normtest

Posted: Sun Apr 22, 2012 1:15 am
by sfarz
Hi,

I would like to use the normtest in a program,
which are the comand references?

Thanks

Re: Normtest

Posted: Mon Apr 23, 2012 8:20 am
by EViews Esther
There is no specific section for the command lines, however, each section has the following information:

Add-in Types: Object of interest
Default Proc Name: Command

For example, if you would like to run from the series, you can use the command:

Code: Select all

series_name.uninorm

as specified in section UniNorm.

Re: Normtest

Posted: Tue Apr 24, 2012 11:42 am
by sfarz
Thanks a lot!
How can i write a program that performs uninorm for many series and store for each series the results?

Re: Normtest

Posted: Tue Apr 24, 2012 1:32 pm
by EViews Esther
Let me introduce one ad hoc approach. After editing the program, UNIFORM.prg (please see the attachment), you can run the following codes:

Code: Select all

group g ser1 ser2 ser3
for !i=1 to g.@count
   %sername = g.@seriesname(!i)
      series {%sername}
     {%sername}.uninorm next
next

Please note this is an ad hoc approach.

Re: Normtest

Posted: Wed Apr 25, 2012 3:13 am
by sfarz
thanks again!