Purpose of PROC DMNEURL
§
29
NOPRINT :
supresses all output printed in the output window.
NPOINT=i :
½
number of discretization points (should be even inbetween 5 and 19). By de-
fault NPOINT= is selected depending on the number of components selected
in the model using the MINCOMP= and MAXCOMP= options.
OPTCRIT=
Ë
Ë
Ï
Ë
Ë
:
specifies the criterion for the optimization:
OPTCRIT=SSE the sum-of-squares error is minimzed.
OPTCRIT=ACC a measure of the accuracy rate is maximized. (For interval
target the Goodman-Kruskal
is applied on a frequency table defined by
deciles of the actual target value.)
OPTCRIT=WSSE a weighted sum-of-squares
criterion is minimzed.
When this option is specified the weight must be specified using the
CRITWGT= option. Currently this option is valid only for binary target.
PALL :
¯
If an OUTSTAT= data set is specified, i.e. only principal components are
being computed, the following table illustrates the output options:
Output
PSHORT
default
PALL
Simple Stat
x
x
x
Eigenvalues
x
x
x
If PMATRIX is specified, the
¼
, the covariance, or the correlation
matrix is also printed (depending on COV and CORR option).
¯
If no OUTSTAT= data set is specified, i.e. a nonlinear model based on
activation and link functions is being optimized, the following table illus-
trates the output options:
Output
NOPRINT
PSHORT
default
PALL
PMATRIX :
This option is valid only if an OUTSTAT= data set is specified, i.e. when
DMNEURL is used only for computing eigenvalues and eigenvectors of the
¼
, covariance, or correlation matrix. If PMATRIX is specified, this matrix
is being printed. Since this matrix may be very large its printout is not included
by that of the PALL option.
POPTHIS :
print the detailed histories of all optimization processes. The PALL option
includes only the summarized forms of the history output (header and result).
PSHORT :
see the PALL option for the amount of output being printed.
PTABLE :
specifies the output of accuracy tables. This option is invoked automatically if
the PALL option is specified.
30
§
PROC DMNEURL: Approximation to PROC NEURAL
SELCRIT=
Ë
Ë
Ï
Ë
Ë
:
specifies the criterion for selecting the best result among all of the activation
functions:
SELCRIT=SSE select solution with smallest sum-of-squares error.
SELCRIT=ACC select solution with largest accuracy rate. (For interval tar-
get the Goodman-Kruskal
is applied on a frequency table defined by
deciles of the actual target value.)
SELCRIT=WSSE select solution with smallest weighted sum-of-squares er-
ror. This option is valid only for binary target. When this option is speci-
fied the weight must be specified using the CRITWGT= option.
SINGULAR=r :
specifies a criterion for the singularity test. The default is
Ö
½
and
should not be changed if there are no significant reasons to do so.
STOPR2=r :
specifies a lower value for the incremental model
Ê
¾
value at which the variable
selection process is stopped. The STOPR2= criterion is used only for the R2
values of the components selected in the range specified by the MINCOMP=
and MAXCOMP= values. The default is
Ö
.
TESTDMDB :
permits the use of a data set generated by PROC DMDB to be specified as a
TESTDATA= input data set. If this option is not specified, the data set specified
with TESTDATA= must be a normal SAS data set.
DECISION Statement
For the syntax of the DECISION statement see the document of PROC DECIDE.
FUNCTION and LINK Statement
An activation function
and a link function
may be specified for the mapping inbe-
tween the component scores
×
and the values
Ý
of the response variable (stage=0)
(or the residuals in stage > 0),
Ý
´
´
µ
´×
µµ
½
Æ
½
Ô
for each activation function
´
µ
½
Ã
. The FUNCTION and LINK state-
ment can be used to specify the functions
´
µ
and :
FUNCTION statement One or more of the following activation functions
can be
specified
Purpose of PROC DMNEURL
§
31
SQUARE
´
·
£
ܵ
£
Ü
TANH
£
Ø
Ò
´
£
ܵ
ARCTAN
£
Ø
Ò´
£
ܵ
LOGIST
ÜÔ´
£
ܵ
´½
·
ÜÔ´
£
ܵ
GAUSS
£
ÜÔ´ ´
£
ܵ
¾
µ
SIN
£
×
Ò´
£
ܵ
COS
£
Ó×´
£
ܵ
EXP
£
ÜÔ´
£
ܵ
If more than one function
´
µ
is specified, each of the specified functions is
evaluated during the estimation process and the best result w.r.t. to the sum-of-
squares residual or accuracy (see SELCRIT= option) is selected. By default all
available activation functions are used.
LINK statement Currently only one of the following link functions can be used for
the outer function :
IDENT
Ü
LOGIST
ÜԴܵ
´½
·
ÜԴܵ
RECIPR
½
Ü
By default, the LOGIST function is used for a binary target and the IDENT(ity)
function is used for interval target. In a parallelized version of PROC DM-
NEURL, multiple functions
could be feasible.
TARGET Statement
TARGET onevar ;
One variable name may be specified identifying the target (response) variable for the
two regressions. Note, that one or more target variables may be specified already
with the PROC DMDB run. If a target is specified in the PROC DMDB run, it must
not be specified in the PROC DMNEURL call.
VAR or VARIABLES Statement
VAR varlist ;
VARIABLES varlist ;
All variables, numeric (interval) and categorical (CLASS) variables which may be
used for independent variables are specified with the VAR statement.
FREQ or FREQUENCY Statement
FREQ onevar ;
FREQUENCY onevar ;
One numeric (interval scaled) variable may be specified as a FREQ variable. Note,
that a rational value is truncated to the next integer. It is recommended to specify
the FREQ variable already in the PROC DMDB run. Then the information is saved
in the catalog and that variable is used automatically as a FREQ variable in PROC
DMNEURL. This also ensures that the FREQ variable is being used automatically
by all other PROCs in the EM project.