PhD Defence Johannes Beyers Louw

Keywords: Causal Inference, Control Variables, Machine Learning, Innovation 
 

"Essays in Innovation and Applied Causal Inference"

This dissertation deals with the methodological rigor in which control variables are applied in management research and investigates innovation diffusion. The former chapters introduce the language of causal diagrams to the management research community, which allows researchers to make their assumptions more explicit and brings transparency to their causal research designs. It also highlights the problematic nature of using correlations to build models that are designed with causality in mind. The final chapter investigates whether a slowdown of knowledge diffusion is the principal cause behind the rise in inequality between firms and combines insights from causal diagrams and recent advances in machine learning to explore the determinants for the diffusion of innovation.  

Click here for the live stream.