Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Statistics›Gamma Regression (GLM)
Regression model

Gamma Regression (GLM)

Gamma Regression (Generalized Linear Model) · Also known as: gamma GLM, gamma generalized linear model, Gamma Regresyonu (GLM)

Gamma regression is a generalized linear model that uses the gamma distribution to model a positive, right-skewed continuous outcome. Developed within the GLM framework of McCullagh and Nelder (1989), it is an alternative to ordinary linear regression for variables such as health-care costs, durations, and income.

ScholarGate
  1. Regression model
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Gamma Regression
Logistic RegressionNegative Binomial Regres…OLS RegressionPoisson RegressionBeta Regression

When to use it

Use gamma regression when the dependent variable is strictly positive (greater than zero) and right-skewed, and the coefficient of variation is roughly constant — that is, the variance grows in proportion to the square of the mean. It suits cross-sectional or panel data with at least about 30 observations and is well suited to skewed continuous outcomes such as costs, durations, and income distributions where normal regression is inappropriate. It is not suitable for outcomes that can be zero or negative.

Strengths & limitations

Strengths
  • Naturally models positive, right-skewed continuous outcomes without log-transforming the response.
  • The log link gives multiplicative, interpretable effects and keeps all fitted means positive.
  • Handles the constant coefficient-of-variation pattern, where variance scales with the square of the mean, better than ordinary least squares.
Limitations
  • Requires a strictly positive outcome; observations equal to zero cannot be modelled directly.
  • Assumes the variance is proportional to the square of the mean (constant coefficient of variation); a different variance pattern misspecifies the model.
  • Estimated by maximum likelihood with no closed form, so it relies on iterative fitting and needs an adequate sample (about 30 or more observations).

Frequently asked

When should I prefer gamma regression over ordinary linear regression?

When the outcome is strictly positive and right-skewed and its variability grows with its mean — typical of costs, durations, and income. Ordinary regression assumes constant variance and can predict negative values, which gamma regression avoids.

Which link function should I use?

The log link is usually preferred because it keeps fitted means positive and makes coefficients act multiplicatively on the outcome. The inverse link is the canonical link for the gamma family but is often harder to interpret.

Can gamma regression handle zeros in the outcome?

No. The gamma distribution is defined only for strictly positive values, so zeros or negative values cannot be modelled directly. A different model, such as a hurdle or two-part model, is needed when zeros occur.

How are the coefficients estimated?

By maximum likelihood. Because there is no closed-form solution, the estimates are obtained iteratively using iteratively reweighted least squares (IRLS), the standard fitting algorithm for generalized linear models.

Sources

  1. McCullagh, P. & Nelder, J. A. (1989). Generalized Linear Models (2nd ed.). Chapman and Hall. DOI: 10.1201/9780203753736 ↗

How to cite this page

ScholarGate. (2026, June 1). Gamma Regression (Generalized Linear Model). ScholarGate. https://scholargate.app/en/statistics/gamma-regression

Related methods

Logistic RegressionNegative Binomial RegressionOLS RegressionPoisson Regression

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Logistic RegressionResearch Statistics↔ compare
  • Negative Binomial RegressionEconometrics↔ compare
  • OLS RegressionEconometrics↔ compare
  • Poisson RegressionEconometrics↔ compare
Compare side by side →

Referenced by

Beta Regression

Similar methods

Generalized Linear ModelBeta RegressionNegative Binomial RegressionPoisson RegressionGAMLSSZero-Inflated Negative Binomial RegressionBayesian Generalized Linear ModelPoisson Rate Regression

Related reference concepts

Logistic RegressionMaximum Likelihood EstimationCox Regression ModelsLogistic DiscriminationNewton-Raphson and Scoring MethodsRegression and Correlation

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Gamma Regression (Gamma Regression (Generalized Linear Model)). Retrieved 2026-07-21 from https://scholargate.app/en/statistics/gamma-regression · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
McCullagh & Nelder (GLM framework)
Year
1989
Type
Generalized linear model
Distribution
Gamma
Link
log (canonical: inverse)
Estimator
Maximum likelihood (IRLS)
Outcome
positive continuous, right-skewed
Related methods
Logistic RegressionNegative Binomial RegressionOLS RegressionPoisson Regression
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account