Causal Mediation
Introduction
Using Parametric and Machine Learning Approaches in R, Virtual workshop, 18-20 Jan 2024
INTEREST CATEGORY: MARKETING RESEARCH
POSTING TYPE: Events
Posted by: Towhidul Islam
A 3-Day Remote Workshop on Causal Mediation
CAUSAL MEDIATION USING PARAMETRIC AND MACHINE LEARNING APPROACHES IN R
Workshop: Thursday, January 18 – Saturday, January 20, 2024.
Visit the link for details:
This workshop will provide the foundations of mediation analysis. Modelling a complex relationship captured by either experimental or observational data, mediation analysis describes the mechanisms and pathways, direct and indirect, by which causal effects operate. Firstly, the widely used traditional model-based Baron and Kenny approach will be introduced. Secondly, the focus will be on the recent developments in causal mediation analysis, notably the counterfactual approach, where effects are defined in a model-free manner using potential outcomes. This causal approach will be implemented with linear and non-linear parametric models and recent advances in machine learning algorithms, specifically with super learners. This workshop will provide insights for policy interventions and the sensitivity of findings to unmeasured confounding variables and measurement errors. Participants will gain an understanding of the concepts, assumptions and limitations of mediation analysis and gain practical tools to implement these techniques with continuous, dichotomous, time-to-event outcomes and mediators.
Live Lecture and Lab Session: 10 am–3:00 pm Eastern Time each day with one hour break at noon. Video-recorded versions will be accessible for 12 weeks,
Payment: The fee of Canadian $650 Regular, $500 for Students.
The workshop is organized by DataOrbit and taught by Towhid Islam, PhD., Professor, Department of Marketing and Consumer Studies, University of Guelph, Canada.