On 12 May 1998 at 11:31:16, David_Bourne (david.aaa.pharm.cpb.uokhsc.edu) sent the message
Hello Stacy Tannenbaum:
Regarding the Winnonlin error message 'subscript out of range', this
means that you must have a do-loop somewhere in your model and the way
you have the programming written it tries to go to a value of n out of
the range of the do-loop. If you say 'do n = 1 to 20' and somewhere
else have 'n = n + 1' then you will end up with a value of 21 for n.
There are only spaces in the matrix for 20 subsripted variables not 21,
so you get the 'subscript out of range'. Hope this is clear.
--
Dan Combs
PK/Metabolism Department
(415) 225-5847 Mailstop 70
Genentech, Inc.
---
From: "Dan Combs"
Date: Mon, 11 May 1998 14:29:09 -0700
reply-to: dzc.-at-.gene.COM
To: PharmPK.at.pharm.cpb.uokhsc.edu
Subject: Re: PharmPK WinNonlin help!
Mime-Version: 1.0
Stacy Tannenbaum:
After more carefully reading your e-mail about the 'subsript out of
range' error message, I see that the error message occurs even using
Winnonlin canned library models. My only suggestion here is that you
have a defective copy of Winnonlin. You might want to contact
Statistical Consultants Inc. for a new copy. Or contact Pharsight Inc.
in Palo Alto, CA - since they now own Winnonlin.
--
Dan Combs
PK/Metabolism Department
(415) 225-5847 Mailstop 70
Genentech, Inc.
---
From: cawello.-a-.schwarzpharma.com
Date: Tue, 12 May 1998 08:33:00 +0100
Sender: cawello.aaa.schwarzpharma.com
Reply-To:
Organization: Schwarz Pharma Group
To: PharmPK.aaa.pharm.cpb.uokhsc.edu
Subject: re: PharmPK WinNonlin help!
Importance: Normal
X-SMF-Hop-Count: 1
MIME-Version: 1.0
Form: Reply
Text: (18 lines follow)
Hi Stacey Tannenbaum,
I think the reason of your problem is the combination of the values of
initial parameters and the step size of the iteration procedure. If you
start the iteration a second time using the same conditions you will find
the same result 'out of range'. If you used initial parameter set calculated
(automatically,..by winnonlin) the next step should be to use the own
initial data set of parameters. Using the procedure with ruler, pencil and
logtransformed concentration vs time data you can 'read' most of this
parameters. Starting the iteration with this parameters you will find a
final result of winnonlin, If there is a realistic connection of your model
and data measured in your study.
Willi Cawello
Humanpharmacology
Schwarz Pharma AG
Monheim am Rhein, Germany
www.cawello.-at-.schwarzpharma.com
---
Date: Tue, 12 May 1998 05:59:35 -0600
From: Vinay Desai
To: PharmPK.-a-.pharm.cpb.uokhsc.edu
Subject: PharmPK WinNonlin help! -Reply
Encoding: 60 Text
Hi Stacey,
I have used WinNonlin extensively and haven't experienced the
frustrations that you seem be having. "Subscript out of Range" is a
Visual Basic error message that is usually evoked when an unhandled
error or exception arises. What operating system are you using? What
version of WinNonlin are you using? If I am not mistaken the current
versions for WinNonlin are ver 1.1 for the standard edition and ver 1.5
for the professional edition. I suspect that the errors you are
encountering are due to mismatched dynamic link libraries (DLL's)
needed by Windows and or WinNonlin or interference from another
application that may be running. Do you have the latest patches from
Microsoft for your operating system- e.g. Service Pack 3 for Windows
NT 4.0. You might also want to reinstall the WinNonlin. Do you have
other applications running while you are modeling? Perhaps, you may
want to shut these apps down.
Hope this helps
Vinay
---
[In a separate message from the original poster I learned that they had
received word from the publisher that a later version of WinNonlin should
fix the problem - db]
Back to the Top