Paper: The Unreasonable Effectiveness of Monte Carlo Simulations in A/B Testing

Introduction

Recently I wrote a paper and published it on arxiv, based on previous articles here on Bytepawn on the usefulness of Monte Carlo simulations in A/B testing:

• Estimating mathematical constants with Monte Carlo simulations
• Early stopping in A/B testing
• A/B testing and the Central Limit Theorem
• Five ways to reduce variance in A/B testing
• A/B testing on social networks
• A/B testing and networks effects
• Bayesian A/B conversion tests
• The Unreasonable Effectiveness of Monte Carlo Simulations in A/B Testing

This paper (pdf, arxiv, source) is a more thorough exploration of these topics.