Comparing Apples and Oranges: Two Examples of the Limits of Statistical Inference, With an Application to Google Advertising Markets
arXiv:1611.10331v1
Originality Synthesis-oriented
AI Analysis
This addresses statistical inference challenges for advertisers and researchers dealing with large-scale online advertising data, but it is incremental as it applies a known bound to a specific domain.
The paper demonstrates that the Cramer-Rao bound imposes fundamental limits on the accuracy of simultaneously estimating values for many Google Ad campaigns, affecting measurement rates in A/B tests.
We show how the classic Cramer-Rao bound limits how accurately one can simultaneously estimate values of a large number of Google Ad campaigns (or similarly limit the measurement rate of many confounding A/B tests).