Using Bootstrapping to Choose Better Predictors

In a review of employee records, management is interested in determining what factors are associated with employee salary increases by fitting a linear model to the difference between current and starting salaries. When bootstrapping a linear model, you can use special resampling methods (residual and wild bootstrap) to obtain more accurate results.

This information is collected in Employee data.sav. See the topic Sample Files for more information.

Note: this example uses the GLM Univariate procedure, and requires the Statistics Base option.

Next