Causal inference using synthetic controls

<p>Last year, Xandr became part of Microsoft, whose mission it is &ldquo;<em>to empower every person and every organization on the planet to achieve more.</em>&rdquo; In order to do that, we need to understand what makes every person and organization successful in the first place. We know that every change&nbsp;<em>we</em>&nbsp;make (to our products or algorithms) and any change&nbsp;<em>our customers</em>&nbsp;make (to their setups or campaigns) can have an effect, be it positive or negative. This is where causal inference comes into play. In this article, I describe how we approach some of these problems using Machine Learning.</p> <p>Unfortunately, conventional Machine Learning methods typically suffer from a critical shortcoming when it comes to causal inference: They are designed to exploit&nbsp;<em>correlations</em>&nbsp;rather than&nbsp;<em>causal relationships</em>. This is often good enough for making predictions, but it is insufficient when it comes to understanding causes and effects (i.e., the&nbsp;<em>why</em>&nbsp;or&nbsp;<em>what if</em>). This is simply because correlation does not imply causation.</p> <p><a href="https://medium.com/data-science-at-microsoft/causal-inference-using-synthetic-controls-d96a890c83a7"><strong>Website</strong></a></p>