values
are sorted by the procedure, and no target value may appear more than once.
Tip:
The DECDATA= data set may be of TYPE=LOSS, PROFIT, or REVENUE.
If unspecified, TYPE=PROFIT is assumed by default. TYPE= is a data set
option that can be specified in parenthesis following the data set name when
the data set is created or when the data set is used.
Options
DECVARS=decision-variable(s)
Specifies the numeric decision variables in the DECDATA= data set that contain the
target-specific consequences for each decision. The decision variables may not contain missing
values. If DECVARS= is not specified, the procedure does not make any decisions or output any
variables that depend on making a decision.
Default:
None
COST=cost-option(s)
Specifies numeric constants that give the cost of a decision, or numeric variables in the DATA=
data set that contain the case-specific costs, or any combination of constants and variables. There
must be the same number of cost constants and variables as there are decision variables in the
DECVARS= option. In the COST= option, you may not use abbreviated variable lists such as
D1-D3, ABC--XYZ, or PQR:. For any case where a cost variable is missing, the results for that
case are set to missing.
Default:
All costs are assumed to be 0.
Note: The COST= option may only be specified when the DECDATA= data set is of
TYPE=REVENUE.
PRIORVAR=variable
Specifies the numeric variable in the DECDATA= data set that contains the prior probabilities to
use for making decisions. Prior probabilities are also used to adjust the total and average profit or
loss. Prior probabilities may not be missing or negative, and there must be at least one positive
prior probability. The priors are not required to sum to 1; if they do not sum to 1, they are
automatically multiplied by a constant to do so. If PRIORVAR= is not specified, no adjustment
for prior probabilities is applied to the posteriors.
Default:
None
OLDPRIORVAR=variable
Specifies the numeric variable in the DECDATA= data set that contains the prior probabilities that
were used when originally fitting the model. The OLDPRIORVAR= option is used only by the
DECIDE procedure. In the DMREG, NEURAL, and SPLIT procedures, the procedure
automatically supplies the values of the old priors if PRIORVAR= is specified.
Note: If OLDPRIORVAR= is specified, PRIORVAR= must also be specified.
Default:
None
Copyright 2000 by SAS Institute Inc., Cary, NC, USA. All rights reserved.
The DECIDE Procedure
FREQ Statement
Specifies a numeric variable whose values represent the frequency of the observation.
Discussion: If one variable in the input data set represents the frequency of occurrence for other
values in the observation, specify the variable's name in a FREQ statement.
Alias: FREQUENCY
FREQ variable;
Required Argument
variable
Specifies a single numeric variable whose value represents the frequency of the observation. If
you use the FREQ statement, PROC DECIDE then treats the data set as if each observation
appeared n times, where n is the value of the FREQ variable for the observation. The FREQ
variable has no effect on decisions or the adjustment for prior probabilities; it only affects the
summary statistics in the OUTFIT= data set. If a value of the FREQ variable is not an integer, the
fractional part is not truncated. If a value of the FREQ variable is less than or equal to 0, the
observation does not contribute to the summary statistics in the OUTFIT= data set, but all of the
variables created in the OUT= data are processed the same way as if the FREQ variable were
positive.
Copyright 2000 by SAS Institute Inc., Cary, NC, USA. All rights reserved.
The DECIDE Procedure
POSTERIORS_Statement__Specifies_which_variables_in_the_DATA=_data_set_contain_the_estimated_posterior_probabilities_that'>POSTERIORS Statement
Specifies which variables in the DATA= data set contain the estimated posterior probabilities that
correspond to the categories of the target variable.
Discussion: The POSTERIORS statement may be specified only with a categorical target variable.
You may not use both a POSTERIORS statement and a PREDICTED statement.
POSTERIORS variable(s);
variable(s)
Specifies the numeric variable(s) in the DATA= data set that contain the estimated posterior
probabilities corresponding to the classes (that is, the categories of the target variable). The results
for a case are set to missing and the P flag is set in the _WARN_ variable for any case where a
posterior probability is missing, negative, or greater than one, or there is a non-0 posterior
corresponding to a 0 prior, or there is not at least one valid positive posterior probability.
CAUTION:
The order of the variables must correspond exactly to the order of the classes in the
DECDATA= data set.
Copyright 2000 by SAS Institute Inc., Cary, NC, USA. All rights reserved.
The DECIDE Procedure
PREDICTED Statement
Specifies which variable in the DATA= data set contains the predicted values of an interval target
variable.
Discussion: The PREDICTED statement may be specified only with an interval target variable. You
may not use both a PREDICTED statement and a POSTERIORS statement.
PREDICTED variable;
variable
Specifies the numeric variable in the DATA= data set that contains the predicted values of an
interval target variable.
Copyright 2000 by SAS Institute Inc., Cary, NC, USA. All rights reserved.