Home - Model selection

Model selection and regularization

A key issue in statistical modeling is to determine the suitable complexity of the model to use. More complex models can capture more of the relationship between the predictors and target variable. But, using more complex models risks over-fitting. We illustrate some simple model order selection techniques.

• Demo: Polynomial order selection with cross-validation
• Lab: Neural decoding motor cortex signals
• Problems: [pdf] [Latex]