Causal Inference on Stopped Random Walks in Online Advertising
This work addresses causal inference for publishers in online advertising, offering a method to estimate effects on revenue and user engagement, but it is incremental as it builds on existing statistical techniques applied to a specific domain.
The paper tackles the problem of estimating long-term treatment effects in online advertising systems, where treatments like auction reserve prices affect user interactions and advertiser budgets, by modeling measurements as stopped random walks and using statistical methods to construct confidence intervals, achieving results with specified confidence levels.
We consider a causal inference problem frequently encountered in online advertising systems, where a publisher (e.g., Instagram, TikTok) interacts repeatedly with human users and advertisers by sporadically displaying to each user an advertisement selected through an auction. Each treatment corresponds to a parameter value of the advertising mechanism (e.g., auction reserve-price), and we want to estimate through experiments the corresponding long-term treatment effect (e.g., annual advertising revenue). In our setting, the treatment affects not only the instantaneous revenue from showing an ad, but also changes each user's interaction-trajectory, and each advertiser's bidding policy -- as the latter is constrained by a finite budget. In particular, each a treatment may even affect the size of the population, since users interact longer with a tolerable advertising mechanism. We drop the classical i.i.d. assumption and model the experiment measurements (e.g., advertising revenue) as a stopped random walk, and use a budget-splitting experimental design, the Anscombe Theorem, a Wald-like equation, and a Central Limit Theorem to construct confidence intervals for the long-term treatment effect.