A Bias Correction Approach for Interference in Ranking Experiments

Online marketplaces use ranking algorithms to determine the rank-ordering of items sold on their websites. The standard practice is to determine the optimal algorithm using A/B tests. We present a theoretical framework to characterize the total average treatment effect (TATE) of a ranking algorithm in an A/B test and show that naive TATE estimates can be biased because of interference. We propose a bias-correction approach that can recover the TATE of a ranking algorithm based on past A/B tests even if those tests suffer from a combination of interference issues. Our solution leverages data across multiple experiments and identifies observations in partial equilibrium in each experiment, that is, items close to their positions under the true counterfactual equilibrium of interest. We apply our framework to data from a travel website and present comprehensive evidence for interference bias in this setting. Next, we use our solution concept to build a customized deep learning model to predict the true TATE of the main algorithm of interest in our data. Counterfactual estimates from our model show that naive TATE estimates of click and booking rates can be biased by as much as 15% and 29%, respectively.

Automated summary

This study provides a theoretical framework for evaluating the effectiveness of ranking algorithms on online marketplaces and identifies potential biases in naive estimates. They propose a method to correct these biases and estimate the total average treatment effect using past A/B test data. The study also uncovers interference bias in data from a travel website and uses a customized deep learning model to accurately estimate the total average treatment effect of the algorithm.