Causality for medical statistics

preview_player
Показать описание
Causality, interaction and observational studies
David Cox,
Nuffield College, Oxford

Causal inference with 0%, 50% and 100% invalid instruments
Jack Bowden,
University of Bristol

Efficacy and mechanisms evaluation using instrumental variables
Richard Emsley,
University of Manchester
Рекомендации по теме
Комментарии
Автор

47:00 Making up mechanism - theory driven, is is all we have...

JCResDoc
Автор

It is so obviously the other way around: when a certain drug is exposed to a certain patient, then the discrete system as a whole, along the markers being measured, enters a different outcome state.

It is just that drugs dont have insurance that we frame everything in this bigoted way.

JCResDoc