arXiv subjects
Publications and source records attributed to Lee Richardson.
Time-varying treatment effects can be estimated with linear regression paired with jackknife confidence intervals. In experiments at a large online video platform, the test based on this estimator is not much less powerful than modern tests for monotonic trends. If you suspect trends in your online experiments, consider using this estimator.
Online platforms prioritize long-term business outcomes, yet typical experiments are far too short to measure these outcomes directly. Our goal in this paper is to collect and share industry knowledge on how to make decisions from short-term experiments that are better aligned with long-term outcomes. Based on a daylong workshop with 26 experts from 15 online platforms and 4 universities, we formulate a series of propositions that reflect current industry knowledge. Participants largely agreed that reversals of sign from short-run to long-run treatment effects are rare, with reversals concentrating in specific cases such as treatments involving content quality signals, hyper-monetization, and pricing. Although the magnitude of treatment effects can shift over time, a "univariate autosurrogate", corresponding to the short-run treatment effect on the long-run metric of interest, is often hard to beat. A recurring theme was the importance of surrogates that are not only (or even primarily) unbiased for true long-run outcomes, but that improve decision-making. Thus, participants generally agreed that simple, interpretable surrogates were generally preferable to elaborate but hard-to-explain surrogate indices. Participants also agreed that, due to concerns about confounding and transportability, experimentally-learned surrogates are generally preferable to observationally-learned surrogates. However, the drawback is that learning good surrogates from experiments typically requires a large, representative portfolio of long-run experiments that few platforms possess. We conclude that there is no substitute for a well-run long-term experiment, whether for learning surrogates or validating them, and we highlight open challenges including evolving treatments, persistent treatments not fully mediated by short-term proxies, and mismatch between experimental samples and the target population.
North star metrics and online experimentation play a central role in how technology companies improve their products. In many practical settings, however, evaluating experiments based on the north star metric directly can be difficult. The two most significant issues are 1) low sensitivity of the north star metric and 2) differences between the short-term and long-term impact on the north star metric. A common solution is to rely on proxy metrics rather than the north star in experiment evaluation and launch decisions. Existing literature on proxy metrics concentrates mainly on the estimation of the long-term impact from short-term experimental data. In this paper, instead, we focus on the trade-off between the estimation of the long-term impact and the sensitivity in the short term. In particular, we propose the Pareto optimal proxy metrics method, which simultaneously optimizes prediction accuracy and sensitivity. In addition, we give an efficient multi-objective optimization algorithm that outperforms standard methods. We applied our methodology to experiments from a large industrial recommendation system, and found proxy metrics that are eight times more sensitive than the north star and consistently moved in the same direction, increasing the velocity and the quality of the decisions to launch new features.
In many randomized experiments, the treatment effect of the long-term metric (i.e. the primary outcome of interest) is often difficult or infeasible to measure. Such long-term metrics are often slow to react to changes and sufficiently noisy they are challenging to faithfully estimate in short-horizon experiments. A common alternative is to measure several short-term proxy metrics in the hope they closely track the long-term metric -- so they can be used to effectively guide decision-making in the near-term. We introduce a new statistical framework to both define and construct an optimal proxy metric for use in a homogeneous population of randomized experiments. Our procedure first reduces the construction of an optimal proxy metric in a given experiment to a portfolio optimization problem which depends on the true latent treatment effects and noise level of experiment under consideration. We then denoise the observed treatment effects of the long-term metric and a set of proxies in a historical corpus of randomized experiments to extract estimates of the latent treatment effects for use in the optimization problem. One key insight derived from our approach is that the optimal proxy metric for a given experiment is not apriori fixed; rather it should depend on the sample size (or effective noise level) of the randomized experiment for which it is deployed. To instantiate and evaluate our framework, we employ our methodology in a large corpus of randomized experiments from an industrial recommendation system and construct proxy metrics that perform favorably relative to several baselines.
Agent-based models (ABMs) simulate interactions between autonomous agents in constrained environments over time. ABMs are often used for modeling the spread of infectious diseases. In order to simulate disease outbreaks or other phenomena, ABMs rely on "synthetic ecosystems," or information about agents and their environments that is representative of the real world. Previous approaches for generating synthetic ecosystems have some limitations: they are not open-source, cannot be adapted to new or updated input data sources, and do not allow for alternative methods for sampling agent characteristics and locations. We introduce a general framework for generating Synthetic Populations and Ecosystems of the World (SPEW), implemented as an open-source R package. SPEW allows researchers to choose from a variety of sampling methods for agent characteristics and locations when generating synthetic ecosystems for any geographic region. SPEW can produce synthetic ecosystems for any agent (e.g. humans, mosquitoes, etc), provided that appropriate data is available. We analyze the accuracy and computational efficiency of SPEW given different sampling methods for agent characteristics and locations and provide a suite of diagnostics to screen our synthetic ecosystems. SPEW has generated over five billion human agents across approximately 100,000 geographic regions in about 70 countries, available online.