Timothy ChaninStatsigLong-live the 95% Confidence IntervalWhy you should use an arbitrary number that has been around as long as modern statistics.Aug 4, 20221Aug 4, 20221
Timothy ChaninStatsigConfused about p-values and hypothesis testing? Let’s play a game.You get to flip a coin and if it’s heads, you win $10. If it’s tails, I win $10. We play twice, tails comes up twice and you owe me $20…Apr 18, 2022Apr 18, 2022
Timothy ChaninStatsigAnnouncing Statsig’s New Pulse Stats EngineToday we’re announcing the launch of Statsig’s new Pulse Stats Engine. Pulse is Statsig’s experimental analysis tool, and our first Stats…Mar 29, 2022Mar 29, 2022
Timothy ChaninStatsigIntroducing Autotune: Statsig’s Multiarmed BanditActual Multi-armed Bandit (MAB) results on the Statsig website. A comparison between Bandit and A/B testing.Feb 3, 2022Feb 3, 2022
Timothy ChaninTowards Data ScienceIn Defense of Zillow’s Besieged Data ScientistsModels Must Be Wielded With CareNov 16, 20216Nov 16, 20216
Timothy ChaninStatsigEmbrace Overlapping A/B Tests and Avoid the Dangers of Isolating ExperimentsAt Statsig, I’ve had the pleasure of meeting many experimentalists from different backgrounds and experiences. How to handle simultaneous…Oct 7, 20214Oct 7, 20214
Timothy ChaninStatsigYou Don’t Need Large Sample Sizes to Run A/B TestsSmall Companies’ Secret Advantage in ExperimentationAug 12, 20211Aug 12, 20211
Timothy ChaninStatsigIntroducing Experiments+We’re launching Experiments+, Statsig’s ridiculously easy-to-use and advanced experimentation (A/B/n testing) tool.Jun 29, 2021Jun 29, 2021
Timothy ChaninStatsigPitfalls of Multi-arm ExperimentsDealing with Significance for Multiple Test GroupsMay 3, 2021May 3, 2021