LogLinkSpec

class pymc_marketing.mmm.link.LogLinkSpec

Log link: E[y] = exp(mu) * target_scale.

When used with LogSaturation, the model becomes a log-log specification where coefficients have an elasticity-like interpretation. Note that under the default scaling pipeline (y_scaled = y / target_scale), the intercept absorbs log(target_scale) and channel data is divided by channel_scale, so the beta coefficients are approximate elasticities with respect to scaled spend rather than strict textbook elasticities with respect to raw spend.

Methods

LogLinkSpec.__init__(*args, **kwargs)
LogLinkSpec.create_media_contribution_deterministic(...)	Register counterfactual total_media_contribution_original_scale and {output_var}_original_scale.
LogLinkSpec.default_intercept(dims)	Return Normal(0, 5) intercept prior (wider for log-scale).
LogLinkSpec.default_likelihood(dims)	Return LogNormal likelihood prior.
LogLinkSpec.inverse_link(mu)	Return exp(mu).
LogLinkSpec.original_scale_transform(...)	Return exp(variable) * target_scale.
LogLinkSpec.validate_likelihood_compatibility(...)	Raise if likelihood is incompatible with link.
LogLinkSpec.validate_target(y)	Raise ValueError if y contains non-positive values.

Attributes

link