VAR, SVAR and SVEC Models: Implementation Within<i>R</i>Package<b>vars</b>

The structure of the package vars and its implementation of vector autoregressive, structural vector Autoregressive and structural vector error correction models are explained in this paper and it is further possible to convert vector error Correction models into their level VAR representation.

Abstract

The structure of the package vars and its implementation of vector autoregressive, structural vector autoregressive and structural vector error correction models are explained in this paper. In addition to the three cornerstone functions VAR(), SVAR() and SVEC() for estimating such models, functions for diagnostic testing, estimation of a restricted models, prediction, causality analysis, impulse response analysis and forecast error variance decomposition are provided too. It is further possible to convert vector error correction models into their level VAR representation. The different methods and functions are elucidated by employing a macroeconomic data set for Canada. However, the focus in this writing is on the implementation part rather than the usage of the tools at hand.