Generalized linear model
A generalized linear model relates a response variable’s distribution to a linear predictor through a link function, typically using a distribution from the exponential family.
Ensemble Methods for Machine Learning
A generalized linear model is a statistical model that combines a linear predictor with a link function and a probability distribution from the exponential family.
- How it is used
- Selects regression variants by matching target labels and distributions within one framework.
- How deep it goes
- Defines the four GLM components and works through label-specific choices, tuning, evaluation, and coefficient interpretation without formal proofs.
- Examples
- Uses bike-rental counts, diabetes diagnosis, and breast-cancer classification to demonstrate distribution choice and coefficient interpretation.
- What you can do after
- Apply it via scikit-learn regressors by selecting distributions, tuning regularization, and evaluating count-prediction models.
Read how Ensemble Methods for Machine Learning explains generalized linear model →
Transfer Learning for Natural Language Processing
A statistical model in which a function links a linear combination of independent variables to the expected value of an output variable.
- How it is used
- Establishes traditional supervised-classification baselines for spam email and movie-review sentiment before transfer learning.
- How deep it goes
- Defines the expected response through a link function over a linear predictor, then applies logistic regression and linear-kernel SVMs.
- Examples
- Uses spam email and movie-review sentiment classification to illustrate applications.
- What you can do after
- Apply logistic regression and linear-kernel SVMs to spam and sentiment classification as traditional baselines.
Read how Transfer Learning for Natural Language Processing explains generalized linear model →
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
A generalized linear model extends linear regression by combining a linear predictor with a response distribution and link function.
- How it is used
- Matches target distributions to tabular prediction problems involving prices, claims, costs, and interruption events.
- How deep it goes
- Compares distribution choices, value constraints, and cross-validated experiments, but does not develop distribution theory.
- Examples
- Uses insurance claims, policy costs, production interruptions, and Airbnb NYC prices as application examples.
- What you can do after
- Apply it via Scikit-learn's TweedieRegressor to compare distribution assumptions for a tabular target.
Also covered in
What to know first
- count dataCount data consist of nonnegative integer-valued observations representing numbers of events, items, or occurrences.
- linear regressionLinear regression is a statistical method that models a numeric response as a linear function of one or more predictor variables with fitted parameters.
- normal distributionA normal distribution is a continuous probability distribution with a symmetric, bell-shaped density determined by its mean and variance.
- logistic regressionLogistic regression is a statistical classification model that estimates class probabilities by applying a logistic or softmax function to a linear combination of input features.
Related concepts
- logistic regressionincludes · Logistic regression is a statistical classification model that estimates class probabilities by applying a logistic or softmax function to a linear combination of input features.
- link functionuses · A link function maps the mean of a response distribution to the linear predictor in a generalized linear model.
- Poisson regressionincludes · Poisson regression is a generalized linear model for count or rate outcomes using a Poisson response distribution and typically a log link.
- glass-box modelkind of · A glass-box model is a predictive model whose structure and decision process are sufficiently transparent for humans to understand.
- linear regressionincludes · Linear regression is a statistical method that models a numeric response as a linear function of one or more predictor variables with fitted parameters.
Questions about generalized linear model
What is generalized linear model?
A generalized linear model relates a response variable’s distribution to a linear predictor through a link function, typically using a distribution from the exponential family.
Which Manning books explain generalized linear model?
Ensemble Methods for Machine Learning, Transfer Learning for Natural Language Processing and Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI.
What should I know before learning generalized linear model?
Manning books that teach generalized linear model usually introduce count data, linear regression, normal distribution and logistic regression first.
What kinds of generalized linear model are there?
Kinds of generalized linear model include logistic regression, Poisson regression, gamma regression and linear regression.
What is generalized linear model a kind of?
Generalized linear model is a kind of glass-box model. A glass-box model is a predictive model whose structure and decision process are sufficiently transparent for humans to understand.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.


