From: Luann Phillips luann.phillips@cognigencorp.com
Subject: [NMusers] Non-Influential Eta Bug
Date: Thu, August 5, 2004 12:18 pm

Hi,

I have a model that is impacted by the non-influential eta bug. I have
been reviewing the example code for the zeta transformation and wondered
if there was a typo.

The example code reads:

ZETA1=ETA(1)+HALF*(ETA(2)+ETA(3))
ZETA2=ETA(2)+HALF*(ETA(1)+ETA(3))
ZETA3=ETA(3)+HALF*(ETA(1)+ETA(2))
CL=THETA(1)*EXP(ZETA1)
V=THETA(2)*EXP(ZETA2)
KA=THETA(3)*EXP(ETA(3))
ALAG=THETA(4)+EXP(ETA(4)

Should the line for KA have EXP(ETA(3)) or EXP(ZETA(3))?
If ETA(3):
Why does KA have ETA instead of ZETA?
Is it the non-influential eta?

Regards,

Luann Phillips
_______________________________________________________

From: "Diane R Mould" drmould@attglobal.net
Subject: RE: [NMusers] Non-Influential Eta Bug
Date: Thu, 5 Aug 2004 12:33:02 -0400

Luann

If Ka was included in the BLOCK structure then its a typo and should be
ZETA

Thanks
Diane 

_______________________________________________________

From: "Ludden, Thomas (MYD)" luddent@iconus.com
Subject: RE: [NMusers] Non-Influential Eta Bug
Date: Thu, 5 Aug 2004 17:21:19 -0400

Luann,
I contacted Stuart Beal and I think the answer is that
KA=THETA(3)*EXP(ETA(3)) should be KA=THETA(3)*EXP(ZETA3).  

Tom

_______________________________________________________