Skip to Content
Date: 9 August 2017
Presenter: Dr Harald van Heerde
Time: 12pm - 1pm
While causality is a hallmark of experimental research, it is much harder to establish causal effects in observational data such as company data, organizational data and surveys. The problem is that the treatment (X variable) is not random, such that its effect on the outcome (Y variable) is often biased. This is one of the thorniest issues in getting empirical research published nowadays. In this talk, Harald van Heerde will discuss new(ish) ways of establishing causal effects in observational data, including instrumental variables, difference-in-difference approaches and Gaussian Copulas.
Page authorised by Web Content Manager
Last updated on Thursday 03 August 2017