The CAUSALTRT Procedure

computes bootstrap-based confidence intervals for the potential outcome means and the treatment effect estimates, and displays the values in the "Analysis of Causal Effect" table. This table includes a column that indicates the number of bootstrap samples used to compute the confidence intervals; this column is not displayed but is available if you save the table as an output data set by using the ODS OUTPUT statement. You can also display this column by modifying the corresponding template.

You can specify one or more of the following types of bootstrap confidence intervals separated by spaces:

The ALPHA= option in the PROC CAUSALTRT statement sets the level of significance that is used to construct the bootstrap confidence intervals. For more information about how the bootstrap-based confidence intervals are computed, see the section Bootstrap Methods. By default, BOOTCI=BC and PROC CAUSALTRT produces bias-corrected confidence intervals based on 1,000 bootstrap samples.

specifies the SAS-data-set that contains the estimates for the potential outcome means and treatment effect for each bootstrap sample that converges.

specifies the number of bootstrap sample data sets (replicates). A maximum of 10,000 bootstrap sample data sets can be requested. By default, NBOOT=1000.

includes in the bootstrap estimation bootstrap samples that have inconsistent class levels in either the treatment or outcome model. By default, for any classification variable in a bootstrap sample data set that does not contain all levels that have been used in fitting either the treatment or outcome model for the original input data set, PROC CAUSALTRT treats the corresponding bootstrap sample estimates as nonconvergent and skips the estimation of the treatment effect. This option overrides the default and continues the estimation of the treatment effect for bootstrap samples that have inconsistent class levels.

produces ODS graphics of bootstrap estimates. By default, if you specify the PLOTS option, a panel is produced with the histograms of the bootstrap estimates for the potential outcome means and treatment effect. No graphics are produced if you specify the PLOTS=NONE option in the PROC CAUSALTRT statement.

ODS Graphics must be enabled before plots can be requested. For example:

ods graphics on;

proc causaltrt method=ipwr;
   model y;
   psmodel trt = x1 x2;
   bootstrap plot;
run;

ods graphics off;

For more information about enabling and disabling ODS Graphics, see the section Enabling and Disabling ODS Graphics in Chapter 24, Statistical Graphics Using ODS.

You can specify the following plot-requests:

You can specify the following global-plot-option:

provides the seed that initializes the random number stream for generating the bootstrap sample data sets (replicates). If you do not specify this option or if you specify a value for n that is less than or equal to 0, the seed is generated from reading the time of day from the computer’s clock. The largest possible value for the seed is .

You can use the SYSRANDOM and SYSRANEND macro variables after a PROC CAUSALTRT step to query the initial and final seed values. However, using the final seed value as the starting seed for a subsequent analysis does not continue the random number stream where the previous analysis ended. The SYSRANEND macro variable provides a mechanism to pass on seed values to ensure that the sequence of random numbers is the same every time you run an entire program. To reproduce the random number stream that was used to generate bootstrap estimates, you must use the same SEED= value and the same number of threads for the analytic computations. When bootstrap resampling is performed, a column that indicates the number of threads used is added to the "Analysis of Causal Effect" table; this column is not displayed but is available if you use the ODS OUTPUT statement to save the table as an output data set. You can also display this column by modifying the corresponding template.