Author Tom Chivers discusses probability, historical origins, Bayesianism, and statistical significance in his book 'Everything is Predictable.' The podcast explores the evolution of probability concepts, historical figures like Pascal and Fermat, and the impact of Bayesianism on scientific statistics.
Understanding Bayes' Theorem and probability misconceptions among MPs highlight the importance of basic statistical concepts.
Historical figures like Bernoulli and Simpson revolutionized statistics by emphasizing sample size and measurement accuracy.
The transformative potential of drug repurposing in medical advancements showcases unexpected benefits and the evolution of treatment strategies.
Deep dives
Bayes' Theorem and Probability Misconceptions
Understanding Bayes' Theorem and its significance in probability, the podcast discusses the misconceptions surrounding probability among MPs. Despite a seemingly simple question about coin tossing, only 52% of MPs answered correctly, revealing gaps in understanding basic probability concepts. The episode highlights the importance of distinguishing between sampling and inferential probabilities in statistical analyses.
Historical Insights and Statistical Advice
The podcast delves into historical figures like Jacob Bernoulli and Thomas Simpson, who made significant contributions to statistics. Bernoulli's work on probability laid the foundation for inferential probabilities, emphasizing the need to consider sample size, error margins, and confidence levels. On the other hand, Simpson provided early statistical advice on aggregating results, focusing on how to enhance measurement accuracy in astronomy.
Serendipitous Medical Discoveries
Exploring the world of medical advancements through drug discoveries, the podcast draws parallels between SGL2 inhibitors and past serendipitous findings in medical treatments. Similar to the unexpected benefits of GLP inhibitors, these drugs designed for diabetes management show promise in addressing various health conditions beyond their initial intended use. The discussion highlights the intriguing pathways of medical progress and the transformative potential of repurposed medications.
Understanding Bayes's Theorem and Its Impact
Bayes's theorem fundamentally emphasizes that any new information or data must be integrated with existing knowledge. This highlights the subjective nature of probability, where statements about likelihood are reflections of one's knowledge and beliefs rather than absolute facts. The example of predicting where a ball lands on a table based on previous results underscores the need to combine prior information with new data for accurate inference. The principle of Bayesianism extends beyond statistics and influences how we perceive and interpret the world around us.
The Evolution of Statistical Methods: Frequentism vs. Bayesianism
The podcast delves into the historical context behind the dominance of frequentist statistical methods over Bayesianism in scientific research. Figures like Ronald Fisher played a crucial role in shaping the statistical landscape, contributing to the relegation of Bayesian inference to a lesser-known niche. Despite Fisher's influence, Bayesianism was independently rediscovered by various researchers like Turing, showcasing its necessity in solving complex inference problems. The debate between subjective Bayesian priors and objective frequentist approaches underscores the ongoing influence and relevance of Bayesian principles in how we approach scientific inquiry.
Everything is Predictable: How Bayes' Remarkable Theorem Explains the World. That’s the new book—out on April 25 in the UK and May 7 in the US—by our very own Tom Chivers!
In this episode of The Studies Show, Tom and Stuart cover some of the historical sections of the book, and talk about where some of our basic ideas about probability come from (it turns out to be a weird combination of inveterate gamblers and Presbyterian ministers).
The Studies Show is sponsored by Works in Progress Magazine - the best place online to find deep discussions of the ideas that have driven human progress, and that might drive it even further in future. The latest issue of Works in Progress is available right now, at worksinprogress.co.
Show notes
* The only citation that matters this week: Tom’s new book, Everything is Predictable. It’s available NOW for pre-order in the UK, and in the US.
* And for those reading this on Substack, here’s the rather lovely front cover:
This is a public episode. If you’d like to discuss this with other subscribers or get access to bonus episodes, visit www.thestudiesshowpod.com/subscribe
Get the Snipd podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share & Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode