Confounder Analysis in Measuring Representation in Product Funnels
This addresses confounder selection in product funnels for platforms like LinkedIn, but it is incremental as it applies an existing method (Shapley values) to a specific causal inference task.
The paper tackled the problem of selecting top confounder variables for coarsened exact matching in causal inference by applying Shapley values, showing they are highly informational and robust for importance-ranking in a LinkedIn dataset.
This paper discusses an application of Shapley values in the causal inference field, specifically on how to select the top confounder variables for coarsened exact matching method in a scalable way. We use a dataset from an observational experiment involving LinkedIn members as a use case to test its applicability, and show that Shapley values are highly informational and can be leveraged for its robust importance-ranking capability.