Abstract Chapter 1 discussed the difficult link between a theory model and an empirical model based on data collected by passive observation. This chapter discusses the VAR approach as a general framework within which statistical inference on macroeconomic hypotheses can be given a valid formulation. An urge to understand more fully why the VAR approach frequently produced results that seemed to cast doubt on conventional theories and beliefs was the motivation for writing this chapter. It focuses on what could be called the economist ‘s approach, as opposed to the statistician ‘s approach, to macroeconomic modelling, distinguishing between models and relations in economics and econometrics.