Quantitude S5E23 A Rosetta Stone for DAGs and SEM
6 snips
Apr 30, 2024 Dive into the intriguing world of structural equation modeling and directed acyclic graphs, with playful insights connecting the two. Discover how linguistic quirks reveal regional identities while exploring concepts like causality and D-separation. Enjoy light-hearted banter about everything from pop culture to childhood memories, all while unraveling the complexities of academic communication. With a dash of humor, this journey turns statistical jargon into relatable talk, making learning both fun and accessible.
AI Snips
Chapters
Transcript
Episode notes
Regional Words Icebreaker
- Patrick Curran and Greg Hancock open with a playful quiz about regional words (pop, soda, garage sale).
- The hosts use regional idioms to illustrate how different communities develop distinct vocabularies.
Rosetta Stone Metaphor
- Patrick recalls seeing the Rosetta Stone in London and explains its role in decoding hieroglyphics.
- He uses that story to motivate a 'Rosetta Stone' for translating between SEM and DAG communities.
SEM As Statistical Causal Statement
- Structural equation modeling encodes hypothesized causal processes as specific statistical statements about moments.
- SEM compares the model-implied moment structure with observed data to evaluate those causal claims.