Online Program Home
My Program

Abstract Details

Activity Number: 663 - Topics in Large-Scale Online Experimentation
Type: Topic Contributed
Date/Time: Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistical Consulting
Abstract #330853 Presentation
Title: Large-Scale Online Experimentation with Quantile Metrics
Author(s): Min Liu* and Xiaohui Sun and Maneesh Varshney and Ya Xu
Companies: LinkedIn Corp. and LinkedIn and LinkedIn Corp. and LinkedIn
Keywords: Controlled experiment; A/B testing; asymptotic distribution; quantile; large-scale computation

Online experimentation (or A/B testing) has been widely adopted in industry as the gold standard for measuring product impacts. Despite the wide adoption, few literatures discuss A/B testing with quantile metrics. Quantile metrics, such as 90th percentile page load time, are crucial to A/B testing as many key performance metrics including site speed and service latency are defined as quantiles. However, with LinkedIn's data size, quantile metric A/B testing is extremely challenging because there is no scalable and accurate variance estimator for the quantile of dependent samples: the bootstrap estimator is accurate, but takes days to compute; the standard asymptotic variance estimate is scalable but results in order-of-magnitude underestimation. In this paper, we present a scalable and accurate methodology for A/B testing with quantiles that is fully generalizable to other A/B testing platforms. It achieves over 500 times speed up compared to bootstrap and has only 2% chance to differ from bootstrap estimates. Beyond methodology, we also share the implementation of a data pipeline using this methodology and insights on pipeline optimization.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program