Split: Inferring Unobserved Event Probabilities for Disentangling Brand-Customer Interactions

TL;DR

This paper introduces a method to infer probabilities of unobserved events in composite data, enabling better disentangling of brand-customer interactions using aggregate data and a modified loss function.

Contribution

It provides a novel identification approach for unobserved event probabilities up to a scalar factor and proposes a practical method to estimate this factor using available aggregate data.

Findings

The method achieves a 46% improvement in average performance.

Validation on synthetic and real data demonstrates effectiveness.

The approach enables inference of unobserved events in complex datasets.

Abstract

Often, data contains only composite events composed of multiple events, some observed and some unobserved. For example, search ad click is observed by a brand, whereas which customers were shown a search ad - an actionable variable - is often not observed. In such cases, inference is not possible on unobserved event. This occurs when a marketing action is taken over earned and paid digital channels. Similar setting arises in numerous datasets where multiple actors interact. One approach is to use the composite event as a proxy for the unobserved event of interest. However, this leads to invalid inference. This paper takes a direct approach whereby an event of interest is identified based on information on the composite event and aggregate data on composite events (e.g. total number of search ads shown). This work contributes to the literature by proving identification of the unobserved events' probabilities up to a scalar factor under mild condition. We propose an approach to identify the scalar factor by using aggregate data that is usually available from earned and paid channels. The factor is identified by adding a loss term to the usual cross-entropy loss. We validate the approach on three synthetic datasets. In addition, the approach is validated on a real marketing problem where some observed events are hidden from the algorithm for validation. The proposed modification to the cross-entropy loss function improves the average performance by 46%.