VECTOR AUTOREGRESSIVE MODELS USING “R”

Abstract

Multivariate data analysis in the context of autoregressive models has evolved as a standard instrument in econometrics. In present, there are developed packages available in R for estimating time series models; one of the most useful package is vars (Pfaff, 2008) containing functions for diagnostic testing, estimation of a restricted models, prediction, causality analysis, impulse response analysis and forecast error variance decomposition. Using the examples provided in the vars vignette, the authors tried to obtain results for the different methods and functions on the base of macroeconomic data set for Romania.

Keywords

• Autoregressive models
• Testing
• vars
• R