liveBook · Manning

Generalization

As explained in Grokking Statistics

Generalization is the ability of a fitted model to perform appropriately on new data beyond the observations used to fit it.

Generalization is a model’s ability to make appropriate predictions on new data rather than merely memorizing the observations used for fitting. section 8.5

A fitted model is not automatically useful: its performance must be checked on data it has not seen. Train/test splits and cross-validation withhold data for later validation. section 8.5 section 1.8

A classification check

A logistic regression model might fit its training observations perfectly by memorizing them. Testing it on withheld observations reveals whether its learned pattern extends beyond the training data. section 8.5

How Grokking Statistics covers it

How deep it goes
Explains memorization, train/test splits, and cross-validation, while deferring advanced logistic-regression metrics and hypothesis tests.
Examples
Uses a fitted logistic regression model, memorized training data, withheld validation data, and cross-validation.
What it assumes
Builds from fitting models and recognizing overfitting before introducing validation on unseen data.
What you can do after
Evaluate whether a fitted model’s predictions extend to unseen data using train/test splits or cross-validation.
How it connects
Connects overfitting, logistic regression, train/test splits, cross-validation, and model evaluation.
Cover of Grokking Statistics
liveBookmanning.com
Read Grokking Statistics on liveBook Get the book at manning.com

Also explained in

Compare how 13 books explain generalization →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Grokking Statistics. The section links point to the parts of the book it draws on.