Validation methods

This subpackage contains all the model validation methods provided by this package. When importing a validation method, be sure to specify the correct module, like so:

from timecave.validation_methods.OOS import Holdout

Out-of-Sample and Prequential methods ensure the training set always precedes the validation set. This is not the case for cross-validation and Markov methods.

Modules

• Base: Contains the class that serves as the basis for every validation method.
• OOS: Implements Out-of-Sample (OOS) methods.
• Prequential: Implements prequential methods (also known as forward validation methods).
• CV: Implements cross-validation (CV) methods.
• Markov: Implements the Markov cross-validation method.
• Weight functions: Provides off-the-shelf weighting functions for use with CV and prequential methods.