BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pydata-london-2026//talk//A38MW7
BEGIN:VTIMEZONE
TZID:GMT
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:GMT
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:BST
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pydata-london-2026-A38MW7@pretalx.com
DTSTART;TZID=GMT:20260606T115000
DTEND;TZID=GMT:20260606T123500
DESCRIPTION:Dynamic Path Analysis (DPA) extends survival analysis with a ca
 usal\, time-varying perspective. This allows causal effects to be decompos
 ed into direct and indirect pathways that evolve over time. The perspectiv
 e is particularly valuable when interventions (exercise) act through media
 tors (weight loss) whose influence changes dynamically in time\, because w
 e get to distil when each driver of our survival probabilities are active 
 and whether their combined effects are harmful or positive. \n\nDespite it
 s conceptual appeal\, DPA remains niche\, with existing implementations li
 mited to frequentist R packages and no Bayesian or Python-native alternati
 ves. In this talk\, I present a Bayesian\, generative implementation of Dy
 namic Path Analysis using PyMC. By discretising time and modelling cumulat
 ive hazard effects with smooth spline priors\, we obtain interpretable tim
 e-varying causal effects with coherent uncertainty quantification. I bench
 mark the approach against canonical dpasurv examples and discuss why DPA f
 ocuses on hazards rather than survival curves.\n\nThis talk is aimed at Py
 thon users interested in survival analysis\, causal inference\, and Bayesi
 an modelling.
DTSTAMP:20260602T223344Z
LOCATION:Hardwick Hub
SUMMARY:Hazards on the Causal Path: Bayesian Time-Varying Survival Analysis
  with PyMC - Nathaniel Forde
URL:https://pretalx.com/pydata-london-2026/talk/A38MW7/
END:VEVENT
END:VCALENDAR