View source on GitHub
|
Implementation of the Model Quality Checks.
Classes
class BaseCheck: A generic, abstract base class for a single, runnable quality check.
class BaseROICheck: Base class for quality checks that rely on ROI priors and posterior estimates.
class BaselineCheck: Checks for negative baseline probability.
class BayesianPPPCheck: Checks for Bayesian Posterior Predictive P-value.
class ConvergenceCheck: Checks for model convergence.
class GoodnessOfFitCheck: Checks for goodness of fit of the model.
class HighVarianceCheck: A check for paid channels with high variance in posterior ROI.
class ImplausibleROICheck: A check for paid channels with implausible posterior ROI estimates.
class PotentialBiasCheck: A check for correlation between paid channels and control variables to flag potential confounding.
class PriorPosteriorShiftCheck: Checks for a significant shift between prior and posterior of ROI.
class ROIConsistencyCheck: Checks if ROI posterior mean is in tails of ROI prior.