{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/c4dd5d6621d54f99aac1739738952f41\" frameborder=\"0\" width=\"1114\" height=\"835\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":835,"width":1114,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":835,"thumbnail_width":1114,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/c4dd5d6621d54f99aac1739738952f41-c77403e3b0e1ad58.gif","duration":440.757,"title":"Lecture 1 - 1.3","description":"In this video, I explore the differences between prediction techniques and causal inference, particularly in the context of decision-making. Using the example of insect repellent and its effect on itchy bumps, I demonstrate how a more accurate predictive model can lead to misleading conclusions about causality. I generated synthetic data showing a true difference of 30% in the likelihood of getting an itchy bump between users and non-users of insect repellent. Interestingly, while the more accurate model predicted a difference of only 4%, the less accurate model correctly identified the true difference. I encourage viewers to consider these nuances when interpreting predictive models and their implications for decision-making."}