Skip to main content

emulators.glm

emulators.glm

Generalized Linear Model (GLM) emulator.

Extends linear regression to non-Gaussian responses via a link function (currently Gaussian/'linear' or 'poisson'), built on statsmodels. Useful when outputs are counts or otherwise not well modeled by plain OLS.

Classes

Name Description
GLM Generalized Linear Model (GLM) emulator.

GLM

emulators.glm.GLM(x=None, y=None, test_fraction=0.25, link='linear')

Generalized Linear Model (GLM) emulator.

Fits a generalized linear model with a configurable link function ('linear' for a Gaussian family, 'poisson' for count data), allowing non-Gaussian output distributions while keeping the model fast and interpretable.

Example

import numpy as np, pandas as pd
from historymatching.emulators.glm import GLM
x = pd.DataFrame({'beta': np.random.rand(40), 'gamma': np.random.rand(40)})
y = pd.DataFrame({'cases': (50 * x['beta']).round()})
em = GLM(x, y, link='poisson')
em.train()
pred = em.predict(x)

Methods

Name Description
get_hyperparameters Return GLM hyperparameters as a JSON-serializable dict.
predict Predict an output using the trained emulator.
print_emulator_description Display detailed specifications (for example, emulator coefficients)
train Fits a Generalised Linear Model.
get_hyperparameters
emulators.glm.GLM.get_hyperparameters()

Return GLM hyperparameters as a JSON-serializable dict.

predict
emulators.glm.GLM.predict(x)

Predict an output using the trained emulator.

Parameters
Name Type Description Default
x pd.DataFrame Input data. Pandas dataframe with columns representing parameter values. required
Returns
Name Type Description
EmulationResults EmulationResults with predicted values and uncertainty intervals.
print_emulator_description
emulators.glm.GLM.print_emulator_description()

Display detailed specifications (for example, emulator coefficients) for the trained emulator.

train
emulators.glm.GLM.train()

Fits a Generalised Linear Model.