Date: Sat, 5 Sep 2009 14:43:01 -0700
Reply-To: Dale McLerran <stringplayer_2@YAHOO.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Dale McLerran <stringplayer_2@YAHOO.COM>
Subject: Re: Fastest Steps for Simulating: Anderson-Darling Goodness of
Fit test for Non-typical distn
In-Reply-To: <6eca73440909050418g10431f71g3541716ddd32a4bf@mail.gmail.com>
Content-Type: text/plain; charset=iso-8859-1
OR Stats,
I've got to be honest with you. The way that you provide
bits and pieces of your problem leaves much to be desired.
The whole picture is never seen. Also, it seems that you
want code handed to you which works right out of the box,
even though you have not provided complete specification
of your problem.
I provided code which computes the Anderson-Darling
statistic. I provided suggestion on how to make that
code suitable for a simulation study. However, you had
only specified some general function G(r.v., s). Note
that at the time that you responded that you wanted to
perform this simulation with G(r.v., s), you had already
received code which implemented the A-D test and that
code employed the variable S as a critical variable in
it's computations. Could it be that there is a conflict
between the value S which is integral to the A-D
statistic computations and the value S which you want
to use for generating a r.v. from the distribution of
interest?
The code which I returned to you was good code up to
some comments that indicated that you needed to revise
those lines because you had not provided complete
specification of your problem. How have you revised
those lines? Might your revisions be causing some
problems?
Again, this brings us back to the very first point that
you only provide bits and pieces of the problem and
then expect the SAS-L community to figure out the
solution for you. We don't know exactly what code you
are executing at this point in time, and so it is not
even possible to provide any specific help to address
your latest issue.
Dale
---------------------------------------
Dale McLerran
Fred Hutchinson Cancer Research Center
mailto: dmclerra@NO_SPAMfhcrc.org
Ph: (206) 667-2926
Fax: (206) 667-5977
---------------------------------------
--- On Sat, 9/5/09, OR Stats <stats112@gmail.com> wrote:
> From: OR Stats <stats112@gmail.com>
> Subject: Re: Fastest Steps for Simulating: Anderson-Darling Goodness of Fit test for Non-typical distn
> To: "Dale McLerran" <stringplayer_2@yahoo.com>
> Cc: SAS-L@listserv.uga.edu
> Date: Saturday, September 5, 2009, 4:18 AM
> It now creates the datasets. But the
> S column is just all zeros and AD column is all -samplesize
> (i.e., -50, -100, -200 etc.)
>
> the error log now is
>
> NOTE: Argument 4 to function CDF at line 3 column 5 is
> invalid.
> NOTE: Argument 4 to function CDF at line 3 column 52 is
> invalid.
>
