Learning Bayesian Statistics cover image

#45 Biostats & Clinical Trial Design, with Frank Harrell

Learning Bayesian Statistics

CHAPTER

Modeling Mistakes - How Do You Solve Them?

When i was in graduate school, we were taught that you could select models on the basis of pea values. I started working with a physician oncolegist, tom rikert, who was running a pattern recognition company using machine learning back in in 19 80. He said do you know, you really need to start holding back data to see if your models are validating? And h i was already getting flat, because we would update our follow up with more patient. After he's pushed me three times, i started doing it. The models were falling apart. So i've been trying to teach everyone that i could get within earshot of that. You know, step

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner