The Intuition behind Vector Auto-Regressions
This video provides an intuitive idea about Vector Auto-Regression models (VARs) and Granger causality.
The practice dataset, R code, and slides that accompany this video are freely available from my github page:
https://github.com/alexcoad/Econometrics
The video starts by moving from regressions, to Auto-Regressions, and then to Vector Auto-Regressions (VARs). A distinction is made between the number of variables in a VAR, and the number of lags in a VAR. Granger causality receives some discussion: presenting the basic idea, and discussing some limitations of Granger causality. Remember: Granger causality is not true causality! Finally, the video discusses an example of a VAR model with 2 variables, using fictional data on firm growth (i.e. sales growth and employment growth).
The video shows some basic equations as well as some code in R software, although viewers can follow the video without any knowledge of R software.