ASReml will automatically delete data records which have a missing value
in the response variable unless the analysis is a
multivariate analysis
or the model term
mv
is fited in the
linear model.
Explicit dropping
!D v
removes records with v or 'missing value' in the field
One of the operators
< <= <> >= >
may be inserted before v to modify the set of records deleted.
If
!D v
is used after
!A
or
!I,
v
should refer to the encoded factor level rather than the value in the
data file.
yield !D <= 0 # Drop records with 0 or negative values.
See Also
!MVREMOVE
to drop records with missing values in design factors