GeneralizedLinearRegression

class causalpy.pymc_models.GeneralizedLinearRegression

Curated generalized linear regression with outcome-scale mu semantics.

Builds eta = X @ beta, applies the configured inverse link to obtain response-scale mu, and wires a family-specific y_hat likelihood.

Supported canonical family/link pairs are Gaussian/identity, Poisson/log, Negative Binomial/log, and Bernoulli/logit. Non-Gaussian families skip Bayesian \(R^2\) scoring because score() returns None.

Parameters:

• family (Literal['gaussian', 'poisson', 'negative_binomial', 'bernoulli']) – Response family. Defaults to "gaussian". The canonical inverse link for the family is exposed read-only via link.

• sample_kwargs (dict[str, Any] | None) – Keyword arguments forwarded to pymc.sample().

• priors (dict[str, Any] | None) – User priors merged over family defaults. Negative Binomial models accept an "alpha" dispersion prior. Custom y_hat overrides are supported only for the Gaussian family.

Methods

GeneralizedLinearRegression.add_coord(name)	Register a dimension coordinate with the model.
GeneralizedLinearRegression.add_coords(coords, *)	Vectorized version of Model.add_coord.
GeneralizedLinearRegression.add_named_variable(var)	Add a random graph variable to the named variables of the model.
GeneralizedLinearRegression.build_model(X, ...)	Define the generalized linear regression PyMC graph.
GeneralizedLinearRegression.check_start_vals(...)	Check that the logp is defined and finite at the starting point.
GeneralizedLinearRegression.compile_d2logp([...])	Compiled log probability density hessian function.
GeneralizedLinearRegression.compile_dlogp([...])	Compiled log probability density gradient function.
GeneralizedLinearRegression.compile_fn(outs, *)	Compiles a PyTensor function.
GeneralizedLinearRegression.compile_logp([...])	Compiled log probability density function.
GeneralizedLinearRegression.copy()	Clone the model.
GeneralizedLinearRegression.create_value_var(...)	Create a TensorVariable that will be used as the random variable's "value" in log-likelihood graphs.
GeneralizedLinearRegression.d2logp([vars, ...])	Hessian of the models log-probability w.r.t.
GeneralizedLinearRegression.debug([point, ...])	Debug model function at point.
GeneralizedLinearRegression.dlogp([vars, ...])	Gradient of the models log-probability w.r.t.
GeneralizedLinearRegression.eval_rv_shapes()	Evaluate shapes of untransformed AND transformed free variables.
GeneralizedLinearRegression.fit(X, y[, coords])	Validate the outcome against the family support, then fit.
GeneralizedLinearRegression.get_context([...])
GeneralizedLinearRegression.initial_point([...])	Compute the initial point of the model.
GeneralizedLinearRegression.logp([vars, ...])	Elemwise log-probability of the model.
GeneralizedLinearRegression.logp_dlogp_function([...])	Compile a PyTensor function that computes logp and gradient.
GeneralizedLinearRegression.make_obs_var(...)	Create a TensorVariable for an observed random variable.
GeneralizedLinearRegression.name_for(name)	Check if name has prefix and adds if needed.
GeneralizedLinearRegression.name_of(name)	Check if name has prefix and deletes if needed.
GeneralizedLinearRegression.point_logps([...])	Compute the log probability of point for all random variables in the model.
GeneralizedLinearRegression.predict(X[, ...])	Predict data given input data X.
GeneralizedLinearRegression.print_coefficients(labels)	Print posterior summaries for regression coefficients and auxiliaries.
GeneralizedLinearRegression.priors_from_data(X, y)	Generate priors dynamically based on the input data.
GeneralizedLinearRegression.profile(outs, *)	Compile and profile a PyTensor function which returns outs and takes values of model vars as a dict as an argument.
GeneralizedLinearRegression.register_data_var(data)	Register a data variable with the model.
GeneralizedLinearRegression.register_rv(...)	Register an (un)observed random variable with the model.
GeneralizedLinearRegression.replace_rvs_by_values(...)	Clone and replace random variables in graphs with their value variables.
GeneralizedLinearRegression.score(X, y[, coords])	Score Bayesian \(R^2\) for Gaussian models; skip for other families.
GeneralizedLinearRegression.set_data(name, ...)	Change the values of a data variable in the model.
GeneralizedLinearRegression.set_dim(name, ...)	Update a mutable dimension.
GeneralizedLinearRegression.set_initval(...)	Set an initial value (strategy) for a random variable.
GeneralizedLinearRegression.shape_from_dims(dims)
GeneralizedLinearRegression.to_graphviz(*[, ...])	Produce a graphviz Digraph from a PyMC model.

Attributes

basic_RVs	List of random variables the model is defined in terms of.
continuous_value_vars	All the continuous value variables in the model.
coords	Coordinate values for model dimensions.
datalogp	PyTensor scalar of log-probability of the observed variables and potential terms.
default_priors
dim_lengths	The symbolic lengths of dimensions in the model.
discrete_value_vars	All the discrete value variables in the model.
family	Curated response family for the built-in GLM.
isroot
link	Read-only canonical inverse link for family.
observedlogp	PyTensor scalar of log-probability of the observed variables.
parent
potentiallogp	PyTensor scalar of log-probability of the Potential terms.
prefix
root
unobserved_RVs	List of all random variables, including deterministic ones.
unobserved_value_vars	List of all random variables (including untransformed projections), as well as deterministics used as inputs and outputs of the model's log-likelihood graph.
value_vars	List of unobserved random variables used as inputs to the model's log-likelihood (which excludes deterministics).
varlogp	PyTensor scalar of log-probability of the unobserved random variables (excluding deterministic).
varlogp_nojac	PyTensor scalar of log-probability of the unobserved random variables (excluding deterministic) without jacobian term.

__init__(family='gaussian', sample_kwargs=None, priors=None)

Parameters:

• sample_kwargs (dict[str, Any] | None) – Dictionary of kwargs that get unpacked and passed to the pymc.sample() function. Defaults to an empty dictionary if None.

• priors (dict[str, Any] | None) – Dictionary of priors for the model. Defaults to None, in which case default priors are used.

• family (Literal['gaussian', 'poisson', 'negative_binomial', 'bernoulli'])

Return type:

None

classmethod __new__(*args, **kwargs)