O'Reilly logo

Practical Predictive Analytics by Ralph Winters

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Model validation

The basic premise of predictive analytics model validation is that you develop your model on one subset of the data (called the training dataset), and then demonstrate that your model has the capability to successfully predict similar resuls on a different set of data. Of primary importance the data set known generically as test. The test dataset is also known as a holdout sample. The training and test datasets are randomly determined before any model building begins, and the data is never changed. The validation data is a third dataset that is sometimes used to further test the validity of the data. It typically contains data the modeler has never seen before and is introduced after one model has been developed and determined ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required