|COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring.|
An Introduction to Dynamic Causal Inference and Multi-state Modelling in Longitudinal Data
If you have a question about this talk, please contact Elena Yudovina.
Longitudinal data are characterised by repeated measurements being taken over time on subjects/units, and in this longitudinal setting it appears natural to consider causality. Where there exists a causal link between two processes or events, the cause must precede its effect. Hence, it seems plausible that a model which aims to uncover a causal relationship should account for the passage of time between cause and effect. I will discuss the idea that causal relationships in longitudinal data can be inferred from an examination of how processes changing over time may influence each another. Multi-state models offer a way of describing dynamic changes in longitudinal data over continuous time and it is through the use of such models, in conjunction with concepts such as composability, local (in)dependence and the well-known Bradford Hill criteria, that I shall argue that evidence for causal relationships in longitudinal data can be provided. Some of these ideas will be demonstrated using data from the University of Toronto Psoriatic Arthritis clinic, concerning damage progression in the hand joints of psoriatic arthritis patients.
This talk is part of the Statistical Laboratory Graduate Seminars series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
Other listsCambridge Ukrainian Studies Lecture Series Millennium Mathematics Project (http://mmp.maths.org) Data Management Roadshow
Other talksA talk by Ameera Patel The annual Breathlessness Research Interest Group Open Lecture Biomarkers for a successful pregnancy Hydrogen–deuterium exchange mass spectrometry Planning for Survival in the Cold War Calories and Corsets: 2000 years of diets and dieting