The SURVEYREG procedure performs regression analysis for sample survey data. The procedure fits linear models and computes regression coefficients and their variance-covariance matrices. PROC SURVEYREG enables you to specify classification effects by using the same syntax that the GLM procedure uses.
PROC SURVEYREG provides hypothesis tests for the model effects. The procedure also provides custom hypothesis tests for linear combinations of the regression parameters. The procedure computes confidence limits for the parameter estimates and also for any specified linear functions of the regression parameters. The procedure can produce an output data set that contains the predicted values from the linear regression, their standard errors and confidence limits, and the residuals.
PROC SURVEYREG also performs regression analysis for domains.
PROC SURVEYREG uses ODS Graphics to create graphs as part of its output. For models that depend on at most one regressor excluding the intercept, the procedure produces fit plots, which can be displayed as bubble plots or heat maps. In bubble plots, the bubble area is proportional to the observation’s weight. In heat maps, the heat color represents the sum of the weights at the corresponding location.
For more information, see Chapter 114: The SURVEYREG Procedure.