The PDLREG Procedure

OUTPUT Statement

OUTPUT OUT= SAS-data-set keyword= option …;

The OUTPUT statement creates an output SAS data set with variables as specified by the following keyword options. See the section “Predicted Values” in Chapter 9: The AUTOREG Procedure, for a description of the associated computations for these options.

ALPHACLI= number

sets the confidence limit size for the estimates of future values of the current realization of the response time series to number, where number is less than one and greater than zero. The resulting confidence interval has 1-number confidence. The default value for number is 0.05, corresponding to a 95% confidence interval.

ALPHACLM= number

sets the confidence limit size for the estimates of the structural or regression part of the model to number, where number is less than one and greater than zero. The resulting confidence interval has 1-number confidence. The default value for number is 0.05, corresponding to a 95% confidence interval.

OUT= SAS-data-set

names the output data.

The following specifications are of the form KEYWORD=names, where KEYWORD= specifies the statistic to include in the output data set and names gives names to the variables that contain the statistics.

CONSTANT= variable

writes the transformed intercept to the output data set.

LCL= name

requests that the lower confidence limit for the predicted value (specified in the PREDICTED= option) be added to the output data set under the name given.

LCLM= name

requests that the lower confidence limit for the structural predicted value (specified in the PREDICTEDM= option) be added to the output data set under the name given.

PREDICTED= name

P= name

stores the predicted values in the output data set under the name given.

PREDICTEDM= name

PM= name

stores the structural predicted values in the output data set under the name given. These values are formed from only the structural part of the model.

RESIDUAL= name

R= name

stores the residuals from the predicted values based on both the structural and time series parts of the model in the output data set under the name given.

RESIDUALM= name

RM= name

requests that the residuals from the structural prediction be given.

TRANSFORM= variables

requests that the specified variables from the input data set be transformed by the autoregressive model and put in the output data set. If you need to reproduce the data suitable for reestimation, you must also transform an intercept variable. To do this, transform a variable that only takes the value 1 or use the CONSTANT= option.

UCL= name

stores the upper confidence limit for the predicted value (specified in the PREDICTED= option) in the output data set under the name given.

UCLM= name

stores the upper confidence limit for the structural predicted value (specified in the PREDICTEDM= option) in the output data set under the name given.

For example, the SAS statements

proc pdlreg data=a;
   model y=x1 x2;
   output out=b p=yhat r=resid;
run;

create an output data set named B. In addition to the input data set variables, the data set B contains the variable YHAT, whose values are predicted values of the dependent variable Y, and RESID, whose values are the residual values of Y.