Predicting Business Revenue with Yelp Consumer Metrics

I just wrote a post for the Yelp Data Blog:

https://www.yelpblog.com/2018/04/predicting-business-revenue-with-yelp-consumer-metrics

Power analysis in A/B Tests

I recently wrote a very basic jupyter notebook demonstrating how to calculate the sample sizes necessary to detect a minimum effect size given desired false positive and false negative rates. I used a web site A/B test as an example, but the technique is generally applicable.