Transforming high skew and kurtosis variables with a multiple Derive node

In this recipe we will create transformations of numeric variables with high skew or kurtosis that makes them more normally distributed with a single Derive node by invoking the Multiple radio button. For many algorithms, normal distributions are assumed and therefore one often transforms variables so that this assumption is more nearly met.

Getting ready

This recipe uses the datafile cup98lrn_reduced_vars3.sav and the stream Recipe - variable construct multipleskew.str.

How to do it...

1. Open the stream Recipe - variable construct multipleskew.str by clicking on File | Open Stream.
2. Make sure the datafile points to the correct path to the datafile cup98lrn_reduced_vars3.sav.
3. Add a Derive node to the stream and connect it to the Type node called First Type.
4. Open the Derive node and click on the Multiple radio button. In the Derive from region, select all monetary variables, including RAMNTALL_2 through TARGET_D. You can select...