meridian.analysis.review.results.PotentialBiasAggregateCases
Stay organized with collections
Save and categorize content based on your preferences.
Cases for Potential Bias Check aggregate result.
Inherits From: ModelCheckCase, BaseCase
Class Variables |
|
NO_CONTROLS
|
<PotentialBiasAggregateCases.NO_CONTROLS: (<Status.REVIEW: 2>, 'No control variables are included in the model. Consider adding control variables to control for potential confounding bias.', 'Some channels have low correlation with all controls. These channels might have biased posterior estimates due to missing potential confounders. We recommend checking if important controls are missing or calibrating these channels using an incrementality experiment to address this.')>
|
|
PASS
|
<PotentialBiasAggregateCases.PASS: (<Status.PASS: 1>, 'All channels have sufficient correlation with control variables.', None)>
|
|
REVIEW
|
<PotentialBiasAggregateCases.REVIEW: (<Status.REVIEW: 2>, "We've detected channel(s) {low_correlation_channels_str} with very low correlation with all included control variables.", 'Some channels have low correlation with all controls. These channels might have biased posterior estimates due to missing potential confounders. We recommend checking if important controls are missing or calibrating these channels using an incrementality experiment to address this.')>
|
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2026-06-17 UTC.
[null,null,["Last updated 2026-06-17 UTC."],[],[]]