This post originally appeared in Ro’s Data Team blog. Stationarity is a crucial property for time series modeling. The problem is, in practice, very few phenomena are actually stationary in their original form. The trick is to employ the right technique for reframing the time series into a stationary form…

Second article in “Measuring Correlation” — Stationarity is a crucial property for time series modeling. The problem is, in practice, very few phenomena are actually stationary in their original form. The trick is to employ the right technique for reframing the time series into a stationary form. One such technique leverages a statistical property called cointegration…

Looker 6.24 introduced conditional alerting for dashboard content. The new feature fills a much-need gap beyond the standard scope of “scheduled looks”, but at first glance appears limited to static threshold alerting. Here we demonstrate how one can leverage the feature’s underlying flexibility to create advanced alerts with self-adaptive thresholds. …

A 5 minute, formula-free refresher on your favorite undervalued tools for data analysis! Choosing the right tool for the job Here are the four main ways to measure correlation, depending on what types of variables you have. This boils down to just three main tests, so the trick is framing both your question and data in a…

Case Study: Cookie Monster’s Monster Cookies — Purpose We’ll implement an end-to-end system that tracks variable costs for different goods sold at a transaction level. The results give a detailed view of variable costs to help with strategic decision making for departments like procurement, supply chain, and inventory management. The files for this case-study can be found here…