<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=648880075207035&amp;ev=PageView&amp;noscript=1">

Free Workshop in Gardena, California:

Seminar_B_2Causal Analysis with Directed Acyclic Graphs and Bayesian Networks

Thursday, January 21, 2016, 1:00 p.m. - 3:00 p.m. 

“…I see no greater impediment to scientific progress than the prevailing practice of focusing all our mathematical resources on probabilistic and statistical inferences while leaving causal considerations to the mercy of intuition and good judgment.” (Pearl, 1999)

Simpson's Paradox

To this day, experiments remain the gold standard for estimating causal effects. However, in many domains and for many reasons, experiments are not feasible. As a result, non-experimental observations are often the only information source we have about a given problem domain. Also, we all know the adage that association does not equal causation.

So, how can we estimate causal effects under these conditions? Unfortunately, no amount of data nor any clever statistical techniques can help us here. Rather, we require (human) causal assumptions about the data-generating process. As it turns out, under the right conditions, a causal effect can sometimes be "identified", in which case the observable association does indeed represent a causal effect.

The objective of this workshop is to provide a remarkably easy-to-use framework for causal identification and estimation using Directed Acyclic Graphs and Bayesian Networks. It will highlight that, in the context of causal inference from observational data, human theoretical reasoning is still crucial. 

This workshop is a "live" version of Chapter 10 in our new book, Bayesian Networks & BayesiaLab (download your free copy here).

Date and Location

Thursday, January 21, 2016
1:00 p.m. - 3:00 p.m.
University of Phoenix
1515 W. 190th St.
Classroom 226
Gardena, CA 90248

Presentation Preview

Workshop Agenda

  • What is Policy Analysis?
  • Causal Inference by Experiment
  • Causal Inference from Observational Data plus Theory
  • Causal Effect Identification
    • Potential Outcomes Framework (Neyman-Rubin Model)
    • Using Directed Acyclic Graphs for Identification (e.g. Back-Door Criterion, etc.)
  • Computing the Effect Size Nonparametrically
    • Using Bayesian Networks and BayesiaLab for Effect Size Computation
      • Pearl's Graph Mutilation
      • Jouffe's Likelihood Matching
  • Managing Uncertainty Probabilistically with Bayesian Networks
    • Uncertain Evidence
    • Uncertainty about Policy Implementation (Probabilistic Intervention)
    • Using BayesiaLab for Optimization under Uncertainty 

Who should attend? 

Policy analysts, decision makers, policy consultants, applied researchers, statisticians, social scientists, data scientists, ecologists, epidemiologists, econometricians, economists, market researchers, knowledge managers, students and teachers in related fields.

About the Presenter

Stefan ConradyStefan Conrady has over 15 years of experience in decision analysis, market research, and product strategy with Fortune 100 companies in North America, Europe, and Asia. Today, in his role as Managing Partner of Bayesia USA and Bayesia Singapore, he is recognized as a thought leader in applying Bayesian networks for research, analytics, and reasoning. In this context, Stefan has recently co-authored a new book, Bayesian Networks & BayesiaLab - A Practical Introduction for Researchers.

Free Registration

Location & Map

University of Phoenix
1515 W. 190th St.
Classroom 226
Gardena, CA 90248

Please keep me posted about upcoming courses and events!