___________________________________________________________________
| |
| MRG_ DATA ITEM |
|_________________________________________________________________|
MEANING: Marginal (MRG_) data item
CONTEXT: $INPUT record and NONMEM data set
USAGE:
$INPUT ... MRG_ ...
DISCUSSION:
MRG_ labels NONMEM's marginal (MRG) data item. The marginal data item
is optional. With a nonobservation record (i.e. MDV=1), it controls
the definition of the value of the PRED item associated with the
record. Basically, 0 means the PRED item is the typical value of F,
and 1 or 2 means it is the expectation of F. (Note that with NM-TRAN,
F is defined by Y.) However, the marginal data item also controls the
definition of the value of any PRED-defined item D that is displayed
in a table or scatterplot (except items stored in the SAVE region) and
associated with the record.
Values are:
0 The PRED item is the typical value of F and a displayed item D is
either the typical value of the variable D or, if conditional
estimates are requested, the conditional estimate of the variable
D.
1 or 2
The values of the PRED and D items are determined by other
things: If the Simulation Step is implemented, and ONLYSIMULATION
is specified, the PRED item is the simulation expectation of F,
over the simulated values of eta for all the individuals in the
data set (including those individuals with no observation
records), and the D item is the simulation expectation of the
variable D. If conditional estimates are obtained (e.g., using
the POSTHOC option), the PRED item is the posterior expectation
of F, over the conditional estimates of eta for all individuals
in the data set (that have some observation records), and the D
item is the posterior expectation of the variable D. Otherwise,
the PRED and D items are as with METHOD=0.
If the expectation is the posterior expectation:
If MRG_=1, the deletion adjustment is not used.
If MRG_=2, the deletion adjustment is used. With the deletion
adjustment, the posterior expectation for a given individual is
obtained by not including in the computed expectation the value
of the variable evaluated at the conditional estimate of his eta.
With MRG_=1 or 2, MDV must also be 1. This use of MDV=1 does not
prevent, as would ordinarily happen, the DV or RES items from being
plotted.
A plot of PRED (and of any PRED-defined item not stored in the SAVE
region) will only include points from records with MRG_=1, unless one
explicitly partitions on MRG_.
If PREDPP is used and NM-TRAN generates MDV, MDV is set to 1 if MRG_
is 1 or 2. If PREDPP is used, NM-TRAN sets MRG_ to 0 (if it is not
already 0) when EVID is not equal to 0.
(See expectation example, expectation block).
REFERENCES: none.
Go to main index.
Created by nmhelp2html v. 1.0 written by Niclas Jonsson (Modified by AJB 5/2006,11/2007)