Causal Inference in Action: Estimating the Impact of External Variables in Retail Using Data Science

Causal inference is a transformative field within data science that delves deep into understanding cause-and-effect relationships between variables. By surpassing mere correlations, causal inference enables us to establish causal relationships, facilitating decision-making and policy evaluation. While randomized control trials (RCTs) serve as the gold standard for causal inference, practical constraints in the retail industry demand the utilization of natural experiments to approximate the effects of external variables. In this presentation, we explore a range of methods, from baseline analysis to Bayesian Networks with tools like CausalNex, that the retail industry employs to estimate causal relationships.

 Jonathan Looman holds a Bachelor's Degree in Civil Engineering and a Masters Degree in Big Data from the University of Stirling. Jonathan kickstarted his data science career at the Shoprite group of companies, where he focused on applying the latest data science techniques to solve retail-related challenges, such as understanding the impact of promotions. He transitioned into the realm of Data Science consulting, collaborating with prominent companies like Alexander Forbes to drive digital innovation. With a genuine passion for leveraging data to gain insights into the world, Jonathan is eager to share his knowledge of data science and practical applications.