Hi all,
I am trying to specify a full OMEGA matrix for simulation purpose. I get issues because some diagonals are zero and some are positive. If I write
$OMEGA BLOCK(3) FIX
0.078
0 0.02
0 0 0
I get this error:
AN ERROR WAS FOUND ON LINE 34 AT THE APPROXIMATE POSITION NOTED:
0 0 0
224 A VARIANCE IS ZERO, BUT THE BLOCK IS NOT FIXED TO ZERO.
I know I could do:
$OMEGA BLOCK(2) FIX
0.078
0 0.02
$OMEGA FIX
0
However, that would be complicated to generate programmatically. Anyway I can tell Nonmem to use a specified OMEGA matrix as is for a simulation?
Thank you,
Philip
Specify full OMEGA matrix for simulation
Looks like a bug.
Workaround is to use a very small diagonal value (1E-30)
Leonid
Quoted reply history
On 7/6/2023 6:01 AM, Philip Harder Delff wrote:
> Hi all,
>
> I am trying to specify a full OMEGA matrix for simulation purpose. I get issues because some diagonals are zero and some are positive. If I write
>
> $OMEGA BLOCK(3) FIX
> 0.078
> 0 0.02
> 0 0 0
>
> I get this error:
> AN ERROR WAS FOUND ON LINE 34 AT THE APPROXIMATE POSITION NOTED:
> 0 0 0
>
> 224 A VARIANCE IS ZERO, BUT THE BLOCK IS NOT FIXED TO ZERO.
>
> I know I could do:
> $OMEGA BLOCK(2) FIX
> 0.078
> 0 0.02
> $OMEGA FIX
> 0
>
> However, that would be complicated to generate programmatically. Anyway I can tell Nonmem to use a specified OMEGA matrix as is for a simulation?
>
> Thank you,
> Philip
Hi Philip,
An exact alternative would be to use this:
$OMEGA BLOCK(2) FIX
0.078
0 0.02
$OMEGA BLOCK(1) FIX
0
Note – I have not tested this but if BLOCK(2) FIX works then BLOCK(1) FIX
should work.
Best wishes,
Nick
--
Nick Holford, Professor Emeritus Clinical Pharmacology, MBChB, FRACP
mobile:NZ+64(21)46 23 53 ; FR+33(6)62 32 46 72
email: [email protected]<mailto:[email protected]>
web: http://holford.fmhs.auckland.ac.nz/
Quoted reply history
From: [email protected] <[email protected]> On Behalf Of
Philip Harder Delff
Sent: Thursday, July 6, 2023 1:10 PM
To: Leonid Gibiansky <[email protected]>; [email protected]
Subject: Re: [NMusers] Specify full OMEGA matrix for simulation
Thanks Leonid, that's pretty enough for me.
Best,
Philip
On July 6, 2023 8:01:41 AM ADT, Leonid Gibiansky
<[email protected]<mailto:[email protected]>> wrote:
Looks like a bug.
Workaround is to use a very small diagonal value (1E-30)
Leonid
On 7/6/2023 6:01 AM, Philip Harder Delff wrote:
Hi all,
I am trying to specify a full OMEGA matrix for simulation purpose. I get issues
because some diagonals are zero and some are positive. If I write
$OMEGA BLOCK(3) FIX
0.078
0 0.02
0 0 0
I get this error:
AN ERROR WAS FOUND ON LINE 34 AT THE APPROXIMATE POSITION NOTED:
0 0 0
224 A VARIANCE IS ZERO, BUT THE BLOCK IS NOT FIXED TO ZERO.
I know I could do:
$OMEGA BLOCK(2) FIX
0.078
0 0.02
$OMEGA FIX
0
However, that would be complicated to generate programmatically. Anyway I can
tell Nonmem to use a specified OMEGA matrix as is for a simulation?
Thank you,
Philip
Thank you Han and Nick,
I agree, splitting the blocks would work too. It's not hard to do, and Leonids
workaround is even easier. I should be able to make it work.
Thank you all for your quick and great suggestions!
Philip
Quoted reply history
On July 6, 2023 10:39:50 PM ADT, "郑晗" <[email protected]> wrote:
>Hi Philip,
>
>In my opinion, each block of the omega matrix should be a positive-definite
>matrix. This is because the omega matrix serves as a variance-covariate
>matrix. However, any block matrix that contains zeros on the diagonal would
>not qualify as a positive-definite matrix.
>
>However, there is one exception to this rule. The block matrix with diagonal
>intentionally fixed to zero allows for the convenient removal of IIV/IOV
>without altering the index of eta/omega. I think splitting the omega matrix
>into two block matrix is a good idea, although it is probally more difficult
>to generate programmatically.
>
>Best wishes,
>
>Han
>
>___________________________
>
>郑晗 博士
>Han Zheng, M.D., Ph.D.
>
>北京博之音科技有限公司 Beijing BioVoice Technology Co., Ltd.
>上海博佳医药科技有限公司 Shanghai BioGuider Medical Technology Co., Ltd.
>地址：上海市 浦东新区 祖冲之路 2277弄世和商务中心 1号楼501室 邮编201203
>Address: 1-501 Shihe Business Center, No. 2277 Zuchongzhi Road, Pudong New
>District, Shanghai 201203
>Mb: 156 1895 5099，E-mail: [email protected]
>
>本邮件（或含附件）中包含法律规定的机密，特许或享有版权的信息，该信息只提供给上述指定的人员或机构使用。如果您非指定收件方，您必须知道任何全部或部分的对本邮件（或附件）进行使用，复制或授权他人使用均是被严格禁止的。如果您误收到此邮件，请告知发件人并退回且在您的系统中删除该邮件所有信息。
>This e-mail with attachments is for use by the intended recipient and contains
>information that may be confidential, privileged or copyrighted under
>applicable law. If you are not the intended recipient, you are here formally
>notified that any use, copying or distribution of this e-mail or attachments,
>in whole or in part, is strictly prohibited. Please notify the sender by
>return and delete the e-mail from your system
>
>On 6 July 2023 at 6:08:14 pm, Philip Harder Delff ([email protected]) wrote:
>
>Hi all,
>
>I am trying to specify a full OMEGA matrix for simulation purpose. I get
>issues because some diagonals are zero and some are positive. If I write
>
>$OMEGA BLOCK(3) FIX
>0.078
>0 0.02
>0 0 0
>
>I get this error:
>AN ERROR WAS FOUND ON LINE 34 AT THE APPROXIMATE POSITION NOTED:
>0 0 0
>
>224 A VARIANCE IS ZERO, BUT THE BLOCK IS NOT FIXED TO ZERO.
>
>I know I could do:
>$OMEGA BLOCK(2) FIX
>0.078
>0 0.02
>$OMEGA FIX
>0
>
>However, that would be complicated to generate programmatically. Anyway
>I can tell Nonmem to use a specified OMEGA matrix as is for a
>simulation?
>
>Thank you,
>Philip
>