Cointegration Modeling: Concepts, ARDL Methods, and Software Advances

Cointegration remains a cornerstone of time-series econometrics which is what led to Engle and Granger receiving the Nobel Prize in economics in 2003. 

In this talk, I will begin by revisiting the issue of spurious regression in non-stationary series and outline the intuition behind cointegration. I will then explore various modeling strategies, with particular emphasis on ARDL (Autoregressive Distributed Lag) models and the bounds testing approach. 

The second part of the presentation will focus on software, where I will introduce the ARDL package in R, compare its functionality with other implementations, including proprietary platforms, and share some thoughts on future enhancements and directions.

Venue: Fulton G20