Skip to content

manalkamal/TMLE_for_breakfast

 
 

Repository files navigation

REDDIE WP4 meeting – LTMLE for breakfast

Welcome

The purpose of this workshop is to gather researchers of REDDIE WP4 who are interested in emulating a diabetes target trial in register data. Specifically, we will work with an augmented version of the R-package ltmle.

The meeting is organized as a mixture of 20 minute talks, a lot of time for discussion, and hands-on programming in R.

Venue

University of Copenhagen Øster Farimagsgade 5 B, DK-1014 Copenhagen K, Denmark Room: CSS 7.0.08

Program

Day 1, May 22, 10 am to 4 pm

10:00-10:05
Hello (Christian Torp-Pedersen)
10:05-10:25
Real randomized trials in diabetes (Bochra Jamal Zareini)
10:25-10:40
Discussion
10:40-11:00
R-coding: Data preparation (I)
11:00-11:20
Coffee break
11:20-11:40
Design of an emulated trial using modern causal inference (Thomas Alexander Gerds)
11:40-12:00
Discussion
12:00-13:00
Lunch break
13:00-13:20
Roadmap of statistical learning, G and Q (Alessandra Meddis)
13:20-13:40
Discussion
13:40-14:15
R-coding: Data preparation (II)
14:15-14:30
Coffee break
14:30-14:50
From register data to covariates, treatment, outcomes (Christian Torp-Pedersen)
14:50-15:10
Discussion
15:10-16:00
R-coding: Data preparation (III)

Day 2, May 23, 9 am to 3 pm

9:00-9:20
Sequential outcome regression (Emilie Wessel Søgaard)
9:20-9:40
Discussion
9:40-10:20
R-coding: LTMLE (I)
10:20-10:35
Coffee break
10:35-10:55
Propensity of treatment, censoring, competing risks (Alessandra Meddis)
10:55-11:40
R-coding: LTMLE (II)
11:40-12:00
Discussion
12:00-13:00
Lunch break
13:00-13:20
LTMLE update step, the fluctuation model (Mark Bech Knudsen)
13:20-13:40
Discussion
13:40-13:55
Coffee break
13:55-14:30
R-coding: LTMLE (III)
14:30-14:50
Super learner: advantages and pitfalls (Thomas Alexander Gerds)
14:50-15:00
Good bye (Christian Torp-Pedersen)

References

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • R 61.3%
  • HTML 38.0%
  • TeX 0.7%