Overview (BOOTSTRAP command)

Bootstrapping is a method for deriving robust estimates of standard errors and confidence intervals for estimates such as the mean, median, proportion, odds ratio, correlation coefficient or regression coefficient. It may also be used for constructing hypothesis tests. Bootstrapping is most useful as an alternative to parametric estimates when the assumptions of those methods are in doubt (as in the case of regression models with heteroscedastic residuals fit to small samples), or where parametric inference is impossible or requires very complicated formulas for the calculation of standard errors (as in the case of computing confidence intervals for the median, quartiles, and other percentiles).

The BOOTSTRAP command signals the beginning of temporary bootstrap samples that are in effect only for the next procedure. See Procedures That Support Bootstrapping for a list of procedures that support bootstrapping.

Options

Resampling method. Simple, stratified, and residuals bootstrap resampling are supported. You can also specify the number of bootstrap samples to take.

Pooling method. Choose between percentile and BCa methods for computing confidence intervals. You can also specify the confidence level.

Basic Specification

The basic specification is the BOOTSTRAP command.

By default, BOOTSTRAP draws 1000 samples using simple bootstrap resampling. When the procedure following BOOTSTRAP is run, the pooling algorithm produces 95% confidence intervals using the percentile method. Since no variables have been specified, no records are excluded from resampling.

Syntax Rules

  • All subcommands are optional.
  • Subcommands may be specified in any order.
  • Only a single instance of each subcommand is allowed.
  • An error occurs if a keyword is specified more than once within a subcommand.
  • Parentheses, equals signs, and slashes shown in the syntax chart are required.
  • The command name, subcommand names, and keywords must be spelled in full.
  • Empty subcommands are not allowed.
  • Any split variable defined on the SPLIT FILE command may not be used on the BOOTSTRAP command.

Limitations

  • BOOTSTRAP does not work with multiply imputed datasets. If there is an Imputation_ variable in the dataset, running BOOTSTRAP will cause an error.
  • BOOTSTRAP should not be used in conjunction with the N OF CASES command.