Date: Tue, 20 May 2003 22:54:41 +0100
Reply-To: John Whittington <John.W@MEDISCIENCE.CO.UK>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: John Whittington <John.W@MEDISCIENCE.CO.UK>
Subject: Re: Constrained Weibull fit
In-Reply-To: <OF32E218DA.C6C83720-ON85256D2C.0064E101@kodak.com>
Content-Type: text/plain; charset="us-ascii"; format=flowed
At 14:26 20/05/03 -0400, john.hixon@kodak.com wrote (in small part):
>Do you have SAS/QC? If so, look at the sample code below.
>Notice that you can specify the threshold, shape, and scale
>parameters for the fitted Weibull distribution. If you do not
>sepcify the parameters then: the threshold parameter defaults to 0,
>and the shape and scale paarmeters are estimated from the data.
>Is this painless enough? :-)
John, no, it's not painless at all - in fact very painful, and at great
risk of putting me onto one of my soapboxes.
... and the reason? No, I don't have SAS/QC and, in response to a couple
of other very nice answers I got privately, no, I don't have SAS/IML
either! The reason for my soapbox is therefore clear. The separation of
IML, ETS and QC from SAS/STAT is, in my opinion, a bit ridiculous; my need
for those other three 'modules' is sufficiently rare that I can't really
justify licensing them all (at appreciable cost), yet a lot of the things
they contain are things that one would expect to be in 'any statistical
package'.
None of that is your fault - so, coming off my soapbox, I'm very grateful
to you for having taken the trouble to provide me which what SHOULD have
been a painless solution!
Kindest Regards
John
----------------------------------------------------------------
Dr John Whittington, Voice: +44 (0) 1296 730225
Mediscience Services Fax: +44 (0) 1296 738893
Twyford Manor, Twyford, E-mail: John.W@mediscience.co.uk
Buckingham MK18 4EL, UK mediscience@compuserve.com
----------------------------------------------------------------
| 
