specifies the transform phases to perform on the input variables. If a request is not specified properly, it is ignored but does not stop other requests from processing. If you do not specify a request at all, the TRANSFORM statement bins the numeric variables from the active table with the BUCKET(5) method.

specifies the transformation request name. The names should be unique among the requests. If you do not specify unique names or do not provide names, then unique names are generated. Specifying a name can be useful because the name is displayed in the output and is used as a prefix to names of the transformed variables.

specifies the input variables to transform. If you do not specify input variables, then all numeric variables in the table are transformed.

specifies how to impute missing values for the input variables. The imputed values can be the mean, median, or a user-specified value.

specifies how to define outlier boundaries and how to treat outliers. The options for defining the outlier-boundaries are either IQR or PERC. The options for defining outlier-treatment are TRIM or WINSOR.

specifies a function for rescaling or standardization to use during the functional transformation phase. The available functions are as follows:

specifies a supervised or unsupervised discretization method. The unsupervised methods determine cutpoints without a target variable.

specifies the event category for binary target variables specified in the TARGET= option. Event categories are needed for the WOE, IV, or GINI index binning evaluation statistics. Multiple event categories are matched in pairs with multiple target variables based on the order they are listed. If more events are listed than target variables, the remaining events are ignored. If more target variables are listed than events, the last event value is used for the remaining target variables.

target=(saleFlag over18Flag) event=('1' '1')

specifies target variables for performing supervised binning or for computing evaluation statistics with the EVALSTATS= option. Multiple target variables are matched in pairs with multiple input variables based on the order the variables are listed. If more target variables are listed than input variables, the remaining target variables are ignored. If more input variables are listed than target variables, the last target variable is used for the remaining input variables.

requests that the server produce SAS scoring code based on the actions that it performed during the analysis. The server generates DATA step code. By default, the code is replayed as an ODS table by the procedure as part of the output of the statement. More frequently, you might want to write the scoring code to an external file by specifying options.

specifies to display the bin details that result from a binning transformation phase.

specifies to score the input variables according to the transformation requests. The scored results and the variables specified in the INPUT=, TARGET=, and FREQ= options are stored in a temporary table. If you want to transfer additional variables, you can specify them with the IDVARS= option.

requests that all variables in the input table are treated as ID variables when a scoring table is produced. In other words, if this option is specified, all variables from the input table, including computed columns, are transferred to the scoring table. This option has no effect unless you specify the SCORE option.

specifies a number between 0 and 1 from which to determine the confidence level when the BIN= option uses the CHIMERGE= method. The default is α = 0.05, which leads to 100 x (1- α)% = 95% confidence limits for the parameter estimates.

specifies how to initialize the bins for supervised binning methods (MDLP, CHIMERGE, or DTREE). The n value for the BUCKET and QUANTILE options is the starting number of bins. The default value is 100.

specifies to place missing values in a separate bin during the binning phase.

avoids the default merging of bins with no observations (empty bins) with their left neighbors. This option is applicable to the MDLP binning method. Empty bins can occur from initializing the number of bins with the BININIT= option set to BUCKET or QUANTILE.

specifies to compute binning evaluation statistics. These statistics are computed from two-way contingency tables between the scored variables and a target variable. The available statistics are as follows:

specifies a numeric variable with a value that represents the frequency of the observation. For example, if the FREQ= variable has the value 5 for a given observation, then that observation represents five observations. The FREQ= option is not available for transform requests that require percentiles in any of the transformation phase. Consequently, if the FREQ= option is specified and percentiles are required, then all such transformation requests are ignored.

specifies the variables from the active table to transfer to the temporary table that is created by scoring the input table. This option has no effect unless the SCORE option is also specified. (See the SCORE option for details about which variables are added to the temporary table by default.) The IDVARS= option should be used to transfer additional columns from the input table to the scoring table.

specifies the convergence tolerance for the iterative algorithm that is used to compute percentiles. Percentiles are calculated in the outlier transformation phase and can also be calculated in the binning transformation phase if quantiles are requested.

specifies the maximum number of iterations in the percentile algorithm. The percentile algorithm is iterative and avoids the cost of copying and sorting the data with multiple passes through the data. You can limit the number of iterations with the PERCMAXITER= option. You can also control the computational demand with the PERCEPSILON= option, which affects the tolerance criterion by which the convergence of the iterative algorithm is judged. If the percentile computations for a particular variable do not converge, the transformation that depends on those percentiles is not performed.

specifies either a quoted string that contains the SAS expression that defines the temporary variables or a file reference to an external file with the SAS statements.

specifies the list of temporary variables for the request. Each temporary variable must be defined through SAS statements that you supply with the TEMPEXPRESS= option.