Causality

Hacking Granger Causality Test with ML ApproachesPhoto by Phoenix Han on UnsplashIn time series forecasting is often helpful to inspect graphically the data at disposal. This helps us understand the dynamics of the phenomena we are analyzing and take decisions accordingly. Despite having a colorful plot with our time series may be fascinating, it may lead to incorrect conclusions. Time series are tricky because often unrelated events may still be visually seen to be related.An example of spurious correlation As rational…

How data scientists approach causal inferencePhoto by Joey Genovese on UnsplashIn a recent Author Spotlight Q&A, Matteo Courthoud reflected on the growing importance of making robust predictions, whether one works in industry or in academia:I think in the future, causal inference will become more and more central and we will see a convergence between the theoretical approach from the social sciences and the data-driven approach from computer science.We hope you read the rest of our lively conversation; in the…

Causality is a broad and complex field. Here’s a map to help you understand it.Map of causality. Image created by the author.The world of causality is broadly split into two main domains:Mainland of causal inference: Causal inference is concerned with understanding the effect of the actions you take. Causal inference provides tools which allow you to isolate and calculate the effect of a change within a system- even if that change never happened in practice. Causal inference can be used to answer the following questions:…

Let’s put ourselves in the shoes of a detective and explore causal inferenceAs we saw in the first part of this two-part article, measuring a causal effect is critical to drawing the right conclusions, because every choice you make or decision you make is usually the result of expected causal relationships.For example:Individual choices:If I go vegan, I’ll reduce my ecological footprint.If I drink this tequila shot, I’ll dance better.Companies:Home-office reduces productivitySpamming users with YouTube Premium Ads will…

If we cannot directly test for causality, what should we do?Image by authorLet me take you with me on a journey in my field of expertise, which is also my passion, my obsession, what I like to call: The science and art of causality.“Causality” refers to the relationship between cause and effect. It’s the idea that one event or action can lead to another event or outcome. In other words, causality is concerned with understanding how things happen and why they happen.In this article we are going to see first going to answer…

How can causality help to advance the state of machine “imagination”?There has been a huge amount of press over the past few months about the incredible advances in text to image modelling. These models can generate human-level artwork based upon a user inputted text prompt and is part of a new wave of state of the art image generation algorithms.Figure 1: Sample images generated by the Stable Diffusion model. Stable Diffusion model outputs licenced under CreativeML Open RAIL-M.It is difficult not to be impressed with the…

Microsoft has recently rolled out a solution for the causality problem made by ML Machine learning is great at extracting patterns out of large amounts of data but not necessarily good at understanding those patterns, especially in terms of what causes them. There are machine learning systems that deal with causality, but so far this has mostly been restricted to research that focuses on small-scale problems rather than practical, real-world systems because it’s been hard to do. On the other hand, deep learning, which is…

Understanding correlation won’t help you with decision making and initiatives measurement in a business setting. A solid grasp of causality is what you need.Photo by Evan Dennis on UnsplashWhether you are a Data Scientist dealing with Decision Science, Marketing, Customer Science, or effective A/B testing, Causal Reasoning is a top skill you should master in your career. Disentangling cause-effect relationships is typically overlooked in business and is a largely poorly understood practice. Many key decisions are taken on…