Causality X Health? What's the biggest challenge in scaling causal inference models? On Monday, join me and my guest, Ehud Karavani (IBM Research), for the premiere of our conversation on causality in health. Ehud will share the experiences and challenges he faced on his causal journey. We will talk about Causallib - an open source Python causal inference library that Ehud created and people who impacted his career. Stay tuned for the premiere link on Monday or subscribe to get the link straight in your inbox: https://bit.ly/3z9rkGa With ❤️ from Causal Bandits Podcast
Causal Inference in health!? I'm looking forward to this one :)
Looking forward!
Research Scientist & Assistant Professor (Practice) at Brown University School of Public Health
5moIn epidemiology, I would say the biggest problem is data quality. For many problems, the benefit provided by flexible estimation methods over simple statistical models (e.g., to estimate nuisance parameters) is marginal. I suspect in many cases it's because the quality of the data is too low and that measurement error becomes the most significant impediment to identification.