Date:         Mon, 16 Sep 2002 10:37:58 -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: missing values
Comments: To: Hollie Rich <hx97403@HOTMAIL.COM>
In-Reply-To:  <F2426beUzmRf6Ci6nx80000009f@hotmail.com>
Content-Type: text/plain; charset=us-ascii

Hollie,

Of course, if you are writing the data out for some other
application which does not recognize SAS missing value constructs,
you must convert the data to something which the other app can
recognize as a missing value.  It is worth heeding the advice of
those who suggest something like -99999999999, which will likely
be far enough out in space that, were a person to forget that they
needed to account for the missing values, then the results of any
analysis would raise a big red flag indicating that something must
be wrong in the data.  We can be honest that one would have
reason to write the data using some converted value but also be
honest that folks do sometimes neglect to indicate the proper
code for missing values for an application, and that it is wise
to protect against such forgetfulness.  Thus, the fully honest
approach to this problem is to accept the advice which both Peter's
offered.  I think this is what all who have weighed in here with
advice would support.  Such advice is not offered with any malice.

BTW, what does the software MTDFREML allow you to do?  From the
name, I would conclude that it performs some REstricted Maximum
Likelihood.  As REML is available in a couple of SAS procedures
(VARCOMP and MIXED), what else does MTDFREML do?

Dale


--- Hollie Rich <hx97403@HOTMAIL.COM> wrote:
> Dear SAS-L,
>
> I need reorganize my data set to create new variable and get starting
> values
> with SAS and will use in different software. This software (MTDFREML)
> does
> not recognize "." as a missing value. -9, -99 and so and can be
> defined as
> missing value for this software. I believe that every researcher
> knows what
> is the honest way to make research and no one has right to give
> advise on
> this issue.


=====
---------------------------------------
Dale McLerran
Fred Hutchinson Cancer Research Center
mailto: dmclerra@fhcrc.org
Ph:  (206) 667-2926
Fax: (206) 667-5977
---------------------------------------

__________________________________________________
Do you Yahoo!?
Yahoo! News - Today's headlines
http://news.yahoo.com