About this document
Ch5 Resampling MethodsNotas by Duglas Ferrera is a document available to read on EtoBox.
The document discusses the differences between training error and test error in statistical learning, emphasizing that training error can underestimate test error. It introduces various methods for estimating test error, including the validation-set approach and K-fold cross-validation, highlighting their advantages and drawbacks. The document also touches on issues related to bias and variance in cross-validation estimates.
- Author
- Duglas Ferrera
- Language
- EN